Creator Talks List

Creator: Voting Village

For years programmers, computer scientists, and cybersecurity experts have known that returning a voted ballot over the internet is fundamentally insecure. Beginning with the 2004 SERVE report, experts have warned of threats including hacking the voter’s device, phishing, man-in-the-middle attacks, spoofing, and attacks on election officials’ systems. New dangers have since emerged, including ransomware attacks.

Yet, 32 states allow some form of internet voting, primarily for overseas military personnel and civilians, though some states also make it available to voters with disabilities and tribal members living on reservations. (For a full list of states that allow internet voting, who is eligible, and what type is permitted, see Internet Voting).

Given the steady stream of reported hacks of government agencies, major corporations, and other institutions, why do so many well-meaning people continue to believe that internet voting will increase voter participation, while ignoring well documented threats?

I will give a brief overview of the history, some of the key forces, and the roles that language (“don’t call it internet voting”), wishful thinking, and money are playing.

People:
    SpeakerBio:  Barbara Simons

An expert on electronic voting, Dr. Barbara Simons has been on the Board of Advisors of the U.S. Election Assistance Commission since was appointed in 2008. She co-authored Broken Ballots: Will Your Vote Count? with Prof. Douglas Jones. She was a member of the National Workshop on Internet Voting that was convened by the National Science Foundation at the request of President Clinton and produced a report on Internet Voting in 2001. She also participated on the Security Peer Review Group for the US Department of Defense’s Internet voting project (SERVE) and co-authored the report that led to the cancellation of SERVE because of security concerns. Simons co-chaired the Association for Computing Machinery (ACM) study of statewide databases of registered voters, she co-authored the League of Women Voters report on election auditing, and she co-authored the July 2015 report of the U.S. Vote Foundation entitled The Future of Voting: End-to-End Verifiable Internet Voting.

Simons was President of ACM, the oldest and largest international educational and scientific society for computing professionals, from 1998 until 2000. She founded ACM’s U.S. Public Policy Committee (now called the U.S. Technology Policy Committee) in 1993 and served for many years as the Chair. She is Board Chair of Verified Voting.

Creator: Blacks In Cyber Village

Join B.I.C. Village for a virtual National Service Panel featuring Capitol Technology University, Blacks in Cybersecurity, and the Military Cyber Professionals Association. This welcoming session will explore national service, mission driven cybersecurity, and the many ways professionals can use their skills to support government, military, academic, and community focused initiatives. Hear from experienced leaders, learn about different career and service pathways, and discover how cybersecurity can create meaningful public impact.

People:
    SpeakerBio:  Carmeshia Miller, BIC Military Relations & Veterans Program Lead and BWIC Curriculum & Learning Experience Lead

Carmeshia �Meshia� Miller is a cybersecurity leader, educator, and advocate with more than 30 years of experience dedicated to expanding access, confidence, and opportunity for military veterans and Black women entering and advancing in tech. She is a retired U.S. Army Warrant Officer, bringing decades of experience in leadership, operations, and risk management to her work in cybersecurity and workforce development. After transitioning from military service, Meshia built a career at the intersection of cybersecurity education, career readiness, governance, risk, and compliance, and community empowerment. She is known for breaking down complex technical concepts into practical, confidence-building learning experiences, especially for those who may feel intimidated by the field. As a leader within Blacks in Cyber and Black Women in Cyber, Meshia focuses on creating safe, affirming spaces where women can learn hands-on technical skills, gain career clarity, and build professional networks that lead to real opportunities. Her work emphasizes not only technical competence but also self-advocacy, resilience, and strategic career navigation in environments where Black women are often underestimated or overlooked. Meshia�s mission is simple but powerful: To ensure Black women don�t just enter cybersecurity�but thrive, lead, and shape its future.

SpeakerBio:  Tiffany Scheurenbrand, Missile Defense Agency (MDA) ICAM Lead

Tiffany Scheurenbrand (Sure-N-Brand) is a cybersecurity leader, U.S. Army veteran, and doctoral student with more than a decade of experience across military, federal, and defense cyber environments. Her background spans cybersecurity operations, information assurance, risk management, secure communications, identity and access management, PKI, privileged access management, Zero Trust, compliance, enterprise security, and team leadership. Throughout her career, Tiffany has supported mission-focused cyber programs in high-pressure environments, leading teams, solving operational security challenges, strengthening governance, and helping organizations improve their overall cyber posture. Her experience includes working across technical, operational, and leadership roles, giving her a broad perspective on how cybersecurity impacts people, mission execution, policy, and national defense. Tiffany is currently pursuing a PhD in Space Cybersecurity, with research focused on Zero Trust policy for secure space commands across space, ground, and cloud environments. She is passionate about national service, mentorship, representation in cybersecurity, and helping the next generation of cyber professionals understand that there is no single path into the field.

SpeakerBio:  Shemeka Chance Jeffrey, G6 Cyber Chief
No BIO available

Creator: Radio Frequency Village

Direction finding used to require expensive, specialized hardware. KrakenSDR changed that. For $749, you get a five-channel coherent software-defined radio that — paired with a custom pentagonal antenna array and open-source signal processing software on a Raspberry Pi — performs real-time passive direction of arrival estimation across 24 MHz to 1766 MHz.

This talk is a practitioner’s account of building and deploying this system across five field runs in Portland, OR. We cover the physics of coherent direction finding and why a single RTL-SDR can’t do it, the hardware build and critical configuration details (including the Pi EEPROM settings that prevent the system from browning out in the field), and the full signal processing pipeline from IQ streams to triangulated fix. We present results from FM broadcast targets used as calibration ground truth, and from a 440 MHz amateur repeater in a dense urban environment where multipath degraded but did not defeat the system.

The barrier to passive RF geolocation is no longer cost. It is knowledge. This talk lowers that barrier and releases the field data, post-processing scripts, and deployment checklist as open source.

People:
    SpeakerBio:  K0YTL

Chris is a second-year PhD student at Portland State University where he teaches network and embedded security by day and points antennas at things by night. His current obsession is passive RF geolocation using multi-channel SDR arrays — specifically, building cheap open-source direction-finding infrastructure with KrakenSDR and a fleet of Raspberry Pis that may or may not be malfunctioning at any given moment. He holds an amateur radio license, has strong opinions about noise floors, and is perpetually one firmware update away from a working system. When not chasing signals through Pacific Northwest parks, he can be found explaining to a classroom why their pentest labs are on fire.

Creator: Red Team Village

This tactic will be a table where hackers can help find and prove 0-day vulnerabilities together at scale. The focus of this tactic is to demonstrate a hands-on methodology to find vulnerabilities, leveraging AI in an effective manner, and proving it has real impact. Then, attendees will search for their own 0-days together across a curated set of software for hunting.

We’ll have a collection of 30+ applications available for immediate analysis and live service deployments to prove out the impact.

I’ll bring a large LED board to track the number of 0-days found as well. This board will be updated in real-time as new vulnerabilities are found.

All 0-days will be responsibly disclosed after DefCon.

People:
    SpeakerBio:  Chris Haller

Chris brings over 15 years of experience in Penetration Testing, Incident Response, Risk Evaluation, Threat Intelligence, and System Administration. While Active Duty, Chris was the Incident Management Lead for the Navy Cyber Defense Operations Command where he specialized in response to attacks on classified and unclassified Navy networks across the globe. Throughout his career, Chris has provided actionable information for stakeholders to make informed decisions about reducing risk to the lowest possible levels, resulting in over 30 CVEs attributed to his work.

Chris has co-authored The Hack is Back: Techniques to Beat Hackers at Their Own Games and has created content on HackTheBox, TryHackMe, and Cybrary. He is an avid CTF player and has recently taken the #1 individual and #1 team position in the National Cyber League, while also operating as the Attack/Defense coach for the US Cyber Team.

Chris is a Principal Security Consultant at Omada Technologies out of Portsmouth, NH. He enjoys helping organizations find and fix complex vulnerabilities before bad actors can use them to cause harm.

Mr. Haller was awarded GIAC Security Expert #329 and has over 30 other certifications.

Creator: Crypto & Privacy Village

Even before the clock struck midnight on January 1st, social media began posting how 2026 is the new 2016. With memes, throwback pictures, and nostalgic posts showing up on TikTok, X, and Instagram. While 2016 conjures up great memories for some, it wasn’t a banner year for cybersecurity. In 2016, several high-profile data breaches took place, including two Yahoo breaches that affected 1.5 billion user accounts. 2016 also saw headlines about incidents at LinkedIn, Oracle, and Dropbox. Commentators even began calling 2016 the “Year of the Data Breach.” If we don’t want 2026 to become the new 2016, we will have to relearn the data privacy lessons from the “Year of the Data Breach.” This presentation will explore the lessons that we should have learned from 2016 by first exploring current cybersecurity trends. Next, we will travel back to 2016 and examine the high-profile incidents that occurred that year. Then we will discuss how the problems we faced in 2016 are still impacting us now. Before, finally outlining ways to apply the lessons from 2016 to make us all a little safer.

People:
    SpeakerBio:  Anthony Hendricks

Anthony Hendricks is a legal problem solver and litigator at Crowe & Dunlevy, one of Oklahoma’s largest and oldest firms. At Crowe & Dunlevy, Anthony serves as founder and chair of the firm’s Cybersecurity and Data Privacy Practice Group. His legal practice focuses on data privacy compliance, regulatory enforcement and permitting, and other “bet-the-company” suits in the areas of data security, privacy, and other complex business litigation. Anthony is an adjunct professor who teaches Cybersecurity Law and Information Privacy courses at Oklahoma City University School of Law. He also hosts “Nothing About You Says Computer Technology,” a weekly podcast on cybersecurity and data privacy viewed through the lens of diverse voices. To learn more about Anthony’s current projects and upcoming speaking events, or listen to the latest episodes of his podcast, visit www.anthonyjhendricks.com

Creator: DCNextGen

The history of hacking is tied to both software and hardware. In both cases, losing the ability to fix and adjust the things you own goes against the core of what we do. For software related projects, you can edit a program or use an open source solution, but for hardware we use 3D printing! In this class you’ll learn about the origins of 3D printing and how everyday tools you have at home can pair with a machine to make the items you’ve purchased on your own!

People:
    SpeakerBio:  Evan
No BIO available

Creator: Adversary Village

Six years ago, Adversary Village started as a simple idea, give the people who study adversaries and adversarial tradecraft a place to call home. What grew out of it was something bigger than any of us expected. A community of offensive security professionals, adversary simulation and emulation specialists, threat researchers, tool builders, philosophers and the endlessly curious, all showing up to share what they know about thinking and acting like the adversary. Year after year the Village has been the place to dig into real adversary tradecraft, offensive TTPs, simulation and emulation techniques, hands-on labs, and the tools that make it all work. This panel looks back at how far we have come and forward at where we are headed. Adversaries never stop evolving, from state-sponsored operators running long, patient campaigns to ransomware crews and everyone in between, and neither can we. We will talk about how the Village has kept pace in a realm that reinvents itself every year, how the craft of adversary simulation and emulation has matured into a discipline of its own, and why faithfully replicating real adversary tradecraft has become essential to testing whether our defenses actually work. We will get into why keeping up is a job no single person or team can do alone, and how a community that trades knowledge freely ends up being the sharpest edge we have got. Most of all, we will celebrate the people who built this, the volunteers, the speakers, the first timers who became regulars. Six years in one thing is clearer than ever! It takes a village to keep up with the adversaries, whether they are nation-state actors, criminal groups, or threats we have not even named yet. Come raise a glass with us and help shape what the next six years look like.

People:
    SpeakerBio:  Abhijith “Abx” B R, Founder of Adversary Village

Abhijith B R, also known by the pseudonym Abx, has more than a decade of experience in the offensive cyber security industry, serves as the Director of BreachSimRange, and Founder of Adversary Village. He is a professional hacker, offensive cyber security specialist, red team consultant, security researcher, trainer and public speaker. Currently, he is building BreachSimRange.io as the Founder and Director and is involved with multiple organizations as a consulting specialist to help them build offensive cyber security operations programs, improve their current security posture, assess cyber defense systems, and bridge the gap between business leadership and security professionals. In the past, he led the offensive security team at Envestnet, Inc., held the position of Deputy Manager – Cyber Security at Nissan Motor Corporation, and prior to that, he worked as a Senior Security Analyst at EY. As the founder of Adversary Village (https://adversaryvillage.org/), Abhijith spearheads a community initiative focused on adversary simulation, adversary-tactics, purple teaming, threat actor/ransomware research-emulation, and offensive cyber security. Adversary Village is part of DEF CON Villages and organizes hacking villages at prominent events such as the DEF CON Hacking Conference, RSA Conference etc. Abx also acts as the Lead of an official DEF CON Group named DC0471. He is actively involved in leading the Tactical Adversary project (https://tacticaladversary.io/), a personal initiative that centers around offensive cyber security, adversary attack simulation and red teaming tradecraft. Abhijith has spoken and delivered trainings at various hacking and cyber security conferences such as, DEF CON hacker convention – Las Vegas, RSA Conference – San Francisco, The Diana Initiative – Las Vegas, DEF CON 28 safemode – DCG Village, Opensource India, Security BSides Las Vegas, BSides San Francisco, BSides Tampa, Hack Space Con – Kennedy space center Florida, Nullcon – Goa, c0c0n – Kerala, BSides Delhi, DEF CON Bahrain, DEF CON Singapore etc.

SpeakerBio:  Bryson Bort, Founder and CEO at SCYTHE

Bryson is the Founder of SCYTHE, a start-up building a next generation attack emulation platform, and GRIMM, a cybersecurity consultancy, and Co-Founder of the ICS Village, a non-profit advancing awareness of industrial control system security. He is a Senior Fellow with the Atlantic Council’s Cyber Statecraft Initiative, the National Security Institute, and an Advisor to the Army Cyber Institute. As a U.S. Army Officer, he served as a Battle Captain and Brigade Engineering Officer in support of Operation Iraqi Freedom before leaving the Army as a Captain. He was recognized as one of the Top 50 in Cyber in 2020 by Business Insider. Bryson received his Bachelor of Science in Computer Science with honors from the United States Military Academy at West Point. He holds a Master’s Degree in Telecommunications Management from the University of Maryland, a Master’s in Business Administration from the University of Florida, and completed graduate studies in Electrical Engineering and Computer Science at the University of Texas.

SpeakerBio:  Anant Shrivastava

Anant Shrivastava is the founder of Cyfinoid Research and a long time offensive security practitioner with a focus on application, cloud, and supply chain security. He has delivered trainings and talks at Black Hat (USA, Europe, Asia), Nullcon, c0c0n, BSides, Rootconf and multiple other events, and runs projects such as Hacking Archives of India to highlight real work from the security community. His courses are built from real consulting and red team experience, with an emphasis on attack chains that actually show up in the field and defenses that teams can implement the next day.

Creator: Bug Bounty Village
Bug bounty has had a weird public story: in the open it looks like a leaderboard and a payout chart, but most of the meaningful change happened behind closed doors in private programs, internal playbooks, and the slow professionalization of triage. We give a clean mental model for what changed, what stayed the same, and which myths keep wasting everyone’s time. With some stories that people may have never heard.

People:
    SpeakerBio:  Chris “flyingtoasters” Holt, Strategic Engagements & Community Architect, Intigriti

Chris Holt, aka flyingtoasters, is a seasoned bug bounty program leader with over 15 years of experience in application and product security. Certified by GAIC, NTISSI, Guinness, PADI, and the USSF, he has spent the last 8 years building and scaling some of the industry’s most respected bug bounty programs. His expertise spans the full spectrum of vulnerability management: from offensive security and researcher engagement to program operations and strategic growth.

SpeakerBio:  Michael Skelton
No BIO available

Creator: AppSec Village

OpenAI designed ChatGPT’s sandbox as a secure runtime with network isolation, strict timeouts, and an AI supervisor filtering commands. Extracting data seemed impossible.

​In this talk, we demonstrate an attack chain shattering this sandbox. Abusing spreadsheet parsing bypasses the supervisor for persistent root execution. We then live-patch the internal Jupyter kernel, hijacking the hidden reasoning channel to execute a Reasoning Injection Attack and extract sensitive data. To exfiltrate it, we bypass isolation by weaponizing the Task Scheduler to launder URLs past web guardrails.

​The attack culminates by exploiting a shared JFrog package manager. We engineered a protocol weaponizing global authentication rate limits, translating lockout timers into a half-duplex covert channel. This provides reliable exfiltration and C&C. Our chain combines file parsing abuse, Chain of Thought hijacking, privilege confusion, and rate limit DoS to orchestrate a C2 network inside ChatGPT.

People:
    SpeakerBio:  Simcha Kosman

Simcha Kosman is a Senior Security Researcher at Palo Alto Networks with over seven years of experience in vulnerability research. He discovered his first vulnerability at age 15, earning his first bug bounty, and has since uncovered security flaws in processors, embedded systems, and large-scale open-source projects. His current work focuses on AI security, exploring the intersection of LLM and software exploitation. Simcha has presented his research in the past at BSides, Nullcon, and Black Hat.

Creator: AI Village

OpenAI designed ChatGPT’s container sandbox as a secure runtime environment, enforcing full network isolation, strict execution timeouts, and an AI supervisor to filter every command. Under this model, owning the container and extracting sensitive data seemed impossible. However, we demonstrate that by chaining file-parsing abuse for persistent execution, reasoning-channel hijacking for data extraction, and shared infrastructure manipulation, an attacker can establish a Cross-tenant data exfiltration.

In this talk, we demonstrate a complete attack chain that shatters ChatGPT’s secure sandbox. By abusing spreadsheet file parsing, we bypass the LLM supervisor to gain persistent, unmonitored root execution. From there, we escalate the attack by live-patching the internal Jupyter kernel to hijack the model’s hidden python.exec reasoning channel, executing a Reasoning Injection Attack to extract sensitive user data. To exfiltrate this data, we bypass network isolation by weaponizing the Task Scheduler to launder malicious URLs past strict web guardrails.

The attack reaches its climax by exploiting a shared JFrog package manager. We engineered a signaling protocol that weaponizes globally visible authentication rate limits, translating these lockout timers into a half-duplex covert channel. This provides reliable data exfiltration and Command and Control from isolated enterprise environments to external attackers. Our exploit chain combines file parsing abuse, Chain of Thought hijacking, privilege confusion, and rate limit Denial of Service to orchestrate a Command and Control (C2) network directly inside ChatGPT.

Breaching AI sandbox agents becomes a critical vulnerability when trust boundaries are shared across millions of users. This research proves that as AI agents gain more capabilities, the attack surface expands dramatically, even when strict security constraints and mitigations are in place.

People:
    SpeakerBio:  Simcha Kosman

Simcha Kosman is a Senior Security Researcher at Palo Alto Networks with over seven years of experience in vulnerability research. He discovered his first vulnerability at age 15, earning his first bug bounty, and has since uncovered security flaws in processors, embedded systems, and large-scale open-source projects. His current work focuses on AI security, exploring the intersection of LLM and software exploitation. Simcha has presented his research in the past at BSides, Nullcon, and Black Hat.

Creator: Mobile Hacking Community
Community Presentation Proposal Title: A Droidwork Orange: A Longitudinal Study of Android Security Research Samuele Doria (University of Padua), Nicholas Miazzo (University of Padova), Tiziano Labruna (Fondazione Bruno Kessler), Eleonora Losiouk (University of Padua) Abstract Android has been the dominant mobile platform for over a decade, but we still lack an understanding of how Android security research has evolved, which problems have received sustained attention, and whether the recent drop in publication volume reflects exhaustion or maturity. This paper presents the first comprehensive longitudinal study of the Android security research ecosystem. We built Ludovico, a curated corpus of 11,691 Android security papers published between 2008 and 2025, complemented by 464 iOS papers and a survey of 86 active researchers. By combining iterative topic modeling with manual validation, analysis of authors’ affiliations and publication venues, and extraction of targeted Android versions, we studied the evolution of topics, contributors, publication venues, and the interplay between platform changes and research priorities. Our results show that the field grew rapidly until a 2016-2018 peak and then contracted, but not because Android security is considered solved: the ecosystem has matured, and advancing the state of the art requires specialized, low-level expertise. We observe a structural transition from App Privacy & Security to Malware Detection, which has dominated the literature since 2017. This is also motivated by the adoption of AI-based techniques: 57.4% of papers focusing on malware

People:
    SpeakerBio:  Nicholas Miazzo, University of Padova

yearly output. We further find that iOS research is more focused on digital forensics, revealing how platform openness shapes research trajectories. Finally, 73.26% of surveyed researchers confirm a real-world impact of Android security research, while highlighting platform hardening, limited access to updated datasets, and funding constraints as the main barriers ahead. Brief Biography Nicholas Miazzo is a Ph.D. student in computer science at the University of Padua, where he completed his master’s degree in 2024. His research now focuses on Android security, particularly leveraging Large Language Model agents to analyze decompiled Android applications and perform taint analysis to automatically detect potential data leakage. He is also a co-author of VirtualPatch [1], a system that uses Android virtualization to deploy security patches independently of device vendors. He is currently contributing to its

SpeakerBio:  Eleonora Losiouk, University of Padua
No BIO available

Creator: Social Engineering Community Village

Large language models (LLMs) can now produce persuasive, personalized phishing content at scale. Yet there is no systematic, reusable benchmark for measuring how this offensive capability varies across models or relates to general capability. We introduce ScamBench: a benchmark and evaluation pipeline for LLM-generated spear phishing. ScamBench contributes (i) a public dataset of over 16,000 personalized phishing emails generated by 20 frontier and open-weight models against 150 synthetic target profiles spanning diverse occupations, life stages, and levels of technical sophistication; (ii) a methodology in which simulated recipients and real human users predict behavioral responses to each email; and (iii) a public leaderboard ranking models by how often their generated emails persuade the target to click a link. Across the 20 models, Claude Opus 4.5 achieves the highest click-through rate at 50.3\%, while even the weakest model elicits clicks at 29.8\%. Together, our findings establish AI-enabled social engineering as an empirically measurable capability that scales with frontier models, with ScamBench providing a reusable benchmark for tracking it as models continue to advance.

People:
    SpeakerBio:  Fred Heiding, Researcher at UC Berkeley’s Center for Long-Term Cybersecurity

Fred Heiding is the executive director of Menlo Park Intelligence and a researcher at UC Berkeley’s Center for Long-Term Cybersecurity. He was previously a doctoral and postdoctoral researcher at the Harvard School of Engineering and Applied Sciences and the Harvard Kennedy School, working with Bruce Schneier and Eric Rosenbach. He is a member of the World Economic Forum’s Centre for Cybersecurity and the Geneva Centre for Security Policy, and serves on the committee of the Harvard and MIT Technology and National Security Conference. He has taught several AI and cybersecurity classes at Harvard, regularly briefs the U.S. Congress on AI-powered cyber threats, and advises foreign governments on national cyber risks. His work has been featured at leading conferences, journals, and media outlets, including The Economist, Reuters, TIME, and Foreign Affairs. Fred has assisted in the discovery of more than 45 critical computer vulnerabilities (CVEs), and he previously made headlines for hacking the King of Sweden and the European Commissioner.

Creator: Cloud Village

We found the same root cause — credentials and destinations resolving independently — across six Azure services spanning AI platform, developer tooling, Kubernetes infrastructure, and traditional enterprise services. User-controlled input reaches an outbound HTTP request carrying Managed Identity tokens, HMAC secrets, or OAuth credentials, and nothing validates where the request goes.

The affected services include Azure AI Foundry (code interpreter and OpenAPI tool), Azure AI Speech, Azure MCP Servers, AKS MCP, and Azure API Management. The impact ranges from stealing live MI tokens for any Azure audience, to dumping production Key Vault URIs and never-expiring write SAS tokens from Microsoft’s own backend, to root RCE on Kubernetes nodes through a readonly tool. One finding has CVE-2026-26118 assigned; the rest are reported to MSRC and pending.

We built attack chains that combine these findings: Foundry MI theft into full tenant reconnaissance, Speech SSRF into persistent backend access via never-expiring SAS, MCP credential relay delivered remotely through prompt injection with no network access to the target, and AKS command injection into cluster-wide lateral movement via Azure’s AKS MCP. The oldest service we found this in — APIM — has been in production for over a decade. The newest — AI Foundry — shipped last year. Same pattern.

The talk includes three live demos, a credential relay taxonomy, and a testing methodology the audience can apply on their next Azure engagement. All findings reported to MSRC before submission.

People:
    SpeakerBio:  Marios Gyftos

Marios Gyftos is a Senior Penetration Tester specializing in cloud security across AWS, Azure, and GCP since 2017. He previously presented at DEF CON 33 Cloud Village on Azure service principal exploitation, BSides Athens 2024, and BSidesChicago 2023 on DevOps attack chains. He is the creator of Azure AppHunter, an open-source tool for scanning Graph API permission misconfigurations. He has reported multiple vulnerabilities to MSRC across Azure AI Foundry, Azure Speech, AKS, Azure MCP, APIM, and Entra ID. Solo speaker — all research, exploitation, and vendor reporting was conducted independently.

SpeakerBio:  Chrysostomos Manousis

Chrysostomos Manousis is a cybersecurity professional and Senior Penetration Tester at Mind The Hack, specializing in penetration testing, red teaming, and the delivery of cybersecurity and information technology services. He holds an MEng in Electrical and Computer Engineering, with a focus on software and hardware engineering. His background spans offensive security across web, cloud, and enterprise environments, and he holds multiple industry-recognized certifications.

Creator: Payment Village

At 50x more authorizations per minute than typical and all for the same amount, alarms go off at issuing banks for a good reason. I’ll share the story of how a card-testing adversary abused a misfeature that checked if a card is valid before vaulting it and how their strategies and my defenses evolved over a week long battle. I will share the techniques I used that ultimately distinguished legitimate customer traffic from automated abuse through residential proxies and creative scripts.

People:
    SpeakerBio:  Levi Schuck, Engineering Manager, Staff Engineer

Took on ownership of security for an engineering org during a data breach, reducing security defects, improving performance and keeping the org PCI-compliant under QSA review. Also builds OAuth/OIDC tooling and is an active contributor in the passkeys ecosystem.

Creator: DCNextGen

Have you ever wanted to make your own electronic badges, but not known where to start? Come listen to a teen BadgeMaker talk about his journey from zero to creating SpaceBadge (DC33) and Clip-Boy (DC34)! Enjoy stories of success and otherwise, all to help you on your path into the BadgeMaking space. Prepare yourself to create the next great badge!

People:
    SpeakerBio:  World Machine
No BIO available

Creator: Ham Radio Village
This talk goes past the Meshtastic getting-started guides and into the advanced builds that push LoRa mesh networks to their limits: bridging MeshCore and Meshtastic so the two ecosystems can talk, rugged repeaters that survive the field, battery technology and enclosures for nodes that run for years, multi-channel concentrators, and high-altitude balloons that turn a handheld into a regional backbone. We’ll also cover the human side, bridging separate mesh communities into one bigger network. Come for the range records, leave knowing how to build a network that outlives the grid.

This talk goes past the Meshtastic getting-started guides and into the advanced builds that push LoRa mesh networks to their limits: bridging MeshCore and Meshtastic so the two ecosystems can talk, rugged repeaters that survive the field, battery technology and enclosures for nodes that run for years, multi-channel concentrators, and high-altitude balloons that turn a handheld into a regional backbone. We’ll also cover the human side, bridging separate mesh communities into one bigger network.

People:
    SpeakerBio:  Eric Escobar, Sophos – Red Team Lead

Eric is a seasoned pentester and a Red Tech Lead at Sophos. On a daily basis he attempts to compromise large enterprise networks to test their physical, human, network and wireless security. He has successfully compromised companies from all sectors of business including: Healthcare, Pharmaceutical, Entertainment, Amusement Parks, Banking, Finance, Technology, Insurance, Military, Retail, Food Distribution, Government, Education, Transportation, Energy and Industrial Manufacturing.

His team consecutively won first place at DEF CON 23, 24, and 25’s Wireless CTF, snagging a black badge along the way. Forcibly retired from competing in the Wireless CTF, he now helps create challenges!

Creator: Aerospace Village

At the Aerospace Village, we strive to Build, Promote, and Inspire our next generation of Aerospace Cybersecurity professionals. To meet this vision, we are proud to sponsor these student research presentations from Cal Poly.

Trusting Linux in Orbit: Process Injection Against CubeSat Flight Software: Dylan Gururajan

Today’s CubeSat flight software increasingly relies on Linux-based frameworks. While this makes development more efficient and flexible, it also imports assumptions from general computing that bring vulnerabilities to these environments.

This talk presents a proof-of-concept attack demonstrating how an attacker, once able to uplink code to a live CubeSat, can leverage standard Linux mechanisms to inject and execute arbitrary code within the primary flight process. By abusing ptrace, dynamic memory allocation, and runtime linking, a malicious library can be loaded directly into a running mission process without exploiting kernel vulnerabilities or binaries on disk.

We walk through the full injection chain, including process discovery, register manipulation, remote syscall staging, and dynamic resolution of libc symbols to invoke dlopen within the target process. We also examine the operational constraints of this technique, including one-shot execution via shared library initialization and implications for attacker tradeoffs.

Software Supply Chain Vulnerabilities in Satellites: Clara Davis

Modern satellite systems rely on complex open-source C/C++ dependency ecosystems that may contain untracked security vulnerabilities but unlike Earth’s software, space systems often cannot be patched after deployment. This research, conducted at Cal Poly, San Luis Obispo, develops a pipeline that extends CCScanner, software dependency analysis tool, for multi-toolchain package manager detection and integrates CNEPS for code clone-based component discovery, with recursive repository cloning, transitive dependency graph creation, and CVE database queries with CVSS severity tracking. By analyzing the full dependency chain rather than direct dependencies, this tool captures vulnerability exposure that other, shallower scans miss. This is a critical gap given that C/C++ projects introduce over 70% of their dependencies implicitly through build systems rather than explicit package managers.

People:
    SpeakerBio:  Clara Davis
No BIO available
SpeakerBio:  Dylan Gururajan
No BIO available

Creator: Aerospace Village

At the Aerospace Village, we strive to Build, Promote, and Inspire our next generation of Aerospace Cybersecurity professionals. To meet this vision, we are proud to sponsor these student research presentations from Embry-Riddle Aeronautical University (ERAU).

Jumpseat Intelligence: Bounded Agentic AI for Aircraft Cyber Defense: Sean McConoughey

Modern commercial aircraft generate continuous streams of system and security logs across a multitude of networked subsystems. These security logs are generally analyzed post-flight at a SOC using rules-based matching that can generate significant false positives. How do you fix that, and what does responsible automation look like in a safety-critical environment where the cost of a wrong decision is measured differently than in a traditional IT context? This talk addresses those questions and presents a working student-developed research prototype that attendees can interact with at the Aerospace Village.

SpaceVE: A Satellite Constellation Cyber Research Environment: Samuil Nikolov

Satellite constellations are increasingly critical infrastructure, supporting navigation, communications, and timing for aviation and a wide range of other domains. Studying the security of those systems at constellation scale presents real challenges for academic researchers working outside of operational environments. SpaceVE is ERAU’s Center for Aerospace Resilient Systems’ answer to that gap: a purpose-built satellite constellation cyber research environment designed to support realistic threat injection, detection research, and challenge scenarios without requiring access to operational spacecraft or proprietary ground systems. Student researchers built SpaceVE from the ground up this summer and will present the architecture, the first research missions, and what the initial findings mean for attacking and defending satellite constellations.

People:
    SpeakerBio:  Sean McConoughey
No BIO available
SpeakerBio:  Samuil Nikolov
No BIO available

Creator: AI Village

Coding agents have the capability to write and ship production code with little human review, which can lead to large-scale security threats. At Corridor, we believe in securing code generation at the source: by augmenting coding agents with the necessary security tooling.

While evaluating our product, we quickly realized that there was no principled way to measure whether a given agent harness actually performs well on the axes we care about (like security). Static benchmarks go stale, get contaminated, and rarely capture the multi-step behavior of an autonomous agent.

To address this, we introduce AgentBreaker, an open-source framework that extends the automated benchmark construction tool AutoBencher, taking it from evaluating single LLM calls to full coding-agent harnesses. We instantiated the framework on our task of secure code generation and show that it reliably surfaces actionable, agent harness-specific weaknesses.

We release AgentBreaker and its methodology so you, too, can evaluate agent harnesses on the axes that matter to you.

People:
    SpeakerBio:  Aditi Narasimhan

Aditi Narasimhan is the founding AI research engineer at Corridor, where she works to improve the security of agentic code generation tools. Previously, she was at Carnegie Mellon, where her research focused on AI ethics, pragmatic theory, and tech ethics pedagogy.

SpeakerBio:  Farzaan Kaiyom

Farzaan has built AI products at startups from the pre-seed stage through Series C, wearing hats as a founder and ML Engineer. He holds a BS/MS in Computer Science from Stanford where he focused on AI safety. While at Stanford CRFM, he worked on HELM, which won the Stanford Open Source Software Prize, and presented research at ICLR.

Creator: Red Team Village

A developer’s MCP configuration exposes a gateway credential. That gateway holds production model keys, an A2A agent can reach it through a delegated tool, and the agent loads instructions from a repository an external contributor can modify. Each component looks harmless in isolation; together, they form a viable path from a low-trust developer environment to production systems. Modern AI infrastructure is full of these hidden chains, spanning MCP servers, agent runtimes, model gateways, local inference hosts, notebooks, and web interfaces. No file declares the entire chain, and no scanner confined to MCP, A2A, or any one AI service can reconstruct it. The most dangerous weakness in the agentic stack is often not a component – it is the trust between components.

AgentHound (https://github.com/adithyan-ak/AgentHound) is an offensive security framework that serves as BloodHound for AI agent infrastructure. Its lean Go collector maps agent clients, MCP servers, A2A agents, tools, resources, identities, credentials, and AI services; a local Neo4j-backed server correlates those observations and derives cross-service reachability to reveal attack paths such as poisoned instructions, credential reuse, impersonation, and potential execution or data exfiltration. What makes it special is its ability to merge evidence from multiple protocols and collectors into one deterministic graph, then rebuild higher-level relationships turning scattered configuration issues into concrete, explainable attack chains across the entire AI-agent ecosystem.

In this live demo, we follow one hidden attack chain end to end — from a foothold on a developer workstation to verified access to a protected, high-value resource. AgentHound discovers the local MCP configuration, connected services, tools, credentials, and resources; correlates them into a walkable attack path; and then safely tests whether that path is genuinely exploitable, upgrading the finding from inferred risk to verified evidence in real time. We finish by poisoning an MCP tool description, detecting the new attack path, and restoring the target from a recovery receipt, showing how AgentHound makes cross-system weaknesses visible, testable, explainable, and reversible.

AgentHound is open source under Apache 2.0. Version 1.0 is released and available now.

People:
    SpeakerBio:  Adithyan Arun Kumar

Adithyan Arun Kumar is an AI Security Engineer at Salesforce, specializing in AI adversarial red teaming, RAG security, and agentic threat modeling across the Agentforce ecosystem. Armed with over five years of full-spectrum offensive security experience and a MS in Information Security from Carnegie Mellon University, he holds advanced certifications including OSEP, OSWE, OSCP, and CRTP. His extensive vulnerability research and offensive tradecraft have earned him hall-of-fame acknowledgments from industry leaders such as Apple, Microsoft, and Intel.

Creator: Red Team Village

AI agents are no longer simple chatbots they autonomously execute code, call external APIs, and make decisions across complex workflows. But what happens when the tools they trust are compromised?

This hands-on workshop explores ASI04: Supply Chain Vulnerabilities from the OWASP Top 10 for Agentic Applications (2026). Participants will attack a deliberately vulnerable AI agent system to understand how adversaries exploit the trust agents place in their dependencies.

Through 10 progressive challenges, attendees will be:

-Install typosquatted packages that exfiltrate secrets on import

-Exploit a trojanized MCP (Model Context Protocol) server masquerading as legitimate tooling

-Execute dependency confusion attacks against “internal” packages

-Discover hidden prompt injections buried in tool descriptions

-Poison RAG knowledge bases to manipulate agent behavior

Each attack demonstrates a real-world vector currently affecting production AI systems. Participants will observe exfiltrated data in real-time on an attacker dashboard, making abstract threats tangible.

Key Takeaways:

Hands-on experience exploiting AI supply chain vulnerabilities

Understanding of OWASP’s agentic AI threat landscape

Practical detection and mitigation strategies

Portable lab environment for continued learning

No prior AI/ML security experience required. Participants leave with the complete lab to continue practicing.

People:
    SpeakerBio:  Aamiruddin Syed

Aamiruddin Syed is a Cybersecurity Professional with over a decade of experience specializing in DevSecOps, Shift-Left Security, Cloud Security, and Internal Penetration Testing. He is the OWASP Agentic AI Supply Chain Project Co-Lead and active contributor to the CSA Agentic AI initiative.

He authored Supply Chain Software Security – AI, IoT, Application Security (Apress/Springer) and has deep expertise in automating security in CI/CD pipelines, infrastructure as code, and cloud hardening. He routinely conducts internal security assessments of critical systems and is known for bridging the gap between security and engineering teams to embed security directly into products.

As recognized advocate for secure development, he is a frequent speaker , delivered workshops and chair sessions at leading industry conferences including RSA Conference, DEFCON, and Black Hat.

—

Author:

1. Fault Detection in Microservice Architectures: Integrating Software Fault Prediction with DevSecOps
2. Supply Chain Software Security: AI, IoT, and Application Security

SpeakerBio:  N A

na

Creator: AppSec Village

AI agents have quickly become a core part of production infrastructure. Autonomous systems that invoke tools, execute code, orchestrate workflows, and interact with external services are now deployed across developer tooling, automation platforms, data pipelines, and enterprise environments at scale.

Yet despite the speed of adoption, almost nothing is known about how these systems are actually built or secured in the wild.

To answer that question, we analyzed more than 86K public repos implementing agent logic across LangChain, LangGraph, CrewAI, AutoGen, and MCP, examining prompt construction, tool implementations, authentication models, and permission boundaries. We paired this with internet-wide infra measurement using Shodan, Censys & ShadowServer, surfacing more than 700K exposed agent-related systems, including Ollama inference servers, Ray clusters, n8n platforms & MCP tool servers, many with little or no authentication and numerous known high-impact CVEs.

People:
    SpeakerBio:  Bar Kaduri

Bar Kaduri is a cybersecurity researcher, leader, and international speaker with over 14 years of experience in cloud security, software supply-chain risk, and emerging AI threats. With hands-on expertise in evaluating and stress-testing AI systems, Bar focuses on building practical, resilient defenses that anticipate risks before they are widely recognized. Bar regularly speaks at industry conferences, delivering clear, research-driven insights that bridge deep technical expertise with strategic security leadership in an AI-driven world

SpeakerBio:  Lidan

Lidan has been programming since childhood, driven by a deep passion for data and AI. He previously served as VP of R&D at SecuredTouch, where he helped pioneer behavioral biometrics. Following the company’s acquisition by Ping Identity, the technology he led became a core component of Ping’s SaaS offering. He later joined Transmit Security, where he built their Identity Threat Detection and Response platform from the ground up, targeting large-tier financial organizations. He scaled the team to more than 30 engineers and data scientists, drove ARR from $0 to $20M, and delivered multiple presentations at leading cybersecurity conferences. In 2025, he co-founded Capsule Security, where he serves as CTO. Capsule addresses one of the most critical challenges in modern cybersecurity: securing AI agents. The company provides end-to-end visibility, risk analysis, threat detection, and runtime protection with advanced detection and response.

Creator: Middle Easterns & Africans in Cyber Security (MEACS)

Your agent has more ways to get owned than you think, and most of them do not look like an attack. Give an AI agent tools, memory, and the ability to install things it found on the internet, and you have built an attack surface nobody has a real security model for yet. This talk maps the whole thing: every place an agentic system can be turned against its user, from the instructions it reads to the marketplaces it trusts, laid out as one picture instead of a pile of scary headlines. Then we get to the uncomfortable part, which is what defending it actually takes. You will leave knowing exactly where the bodies are buried, whether you build agents or defend them.

People:
    SpeakerBio:  Amber Bennoui
No BIO available

Creator: Blacks In Cyber Village

Join us for a deep dive into building an AI-powered framework for threat investigation! Security operations often grapple with fragmented data and alert overload. This session moves beyond basic AI summarization to explore how to design and implement tool-enabled AI agents that actively investigate security alerts. Discover how to create structured agents with controlled access to your internal data and external threat intelligence, enabling them to gather context, reason through findings, and produce actionable summaries. We ll cover key architectural considerations, guardrails, and human-in-the-loop controls. A live demo will showcase the full workflow, providing attendees with a clear blueprint for implementing responsible, AI-assisted threat investigation in real-world SOC environments.

People:
    SpeakerBio:  Samson Adewale, Cybersecurity Professional / Community Speaker

Samson Adewale is a Senior Cybersecurity Engineer specializing in Cloud Security, DevSecOps, and Cloud-Native architectures. They design secure platforms, automate incident response, and build tooling for high-velocity teams. As an AI enthusiast, Samson is passionate about applying AI and LLM technologies to enhance security operations and streamline cybersecurity workflows.

Creator: DCNextGen

Wondering how hacking with AI takes place? Join us to find out along with tips on staying safe with some fun and laughs along the way! No prereqs necessary, just curiosity!

People:
    SpeakerBio:  Zoe Reid+Echo419
No BIO available

Creator: Bug Bounty Village
AI is reshaping bug bounty from both sides: how hunters use it to find more bugs, and the brand-new bug classes it creates in production. Built on Hazem Elsayed’s (@hacktus) research and presented by Ciarán Cotter (@monke), who delivers the talk on stage and brings additional cases from his own AI security work.

People:
    SpeakerBio:  Ciarán “monke” Cotter

Ciarán, also known as “monke”, is a full-time bug bounty hunter and founder of Starstrike, an AI security startup. He is a Google AI Bugswat MVH and has won 2nd Place twice. Ciarán is also an active member of the BT6 AI hacking collective organised by Pliny the Liberator.

Creator: Blue Team Village

People:
    SpeakerBio:  Anirudh Pratap Singh

Anirudh Pratap Singh works in IT security and support, with hands-on experience across cybersecurity operations, infrastructure, cloud systems, and user support. He holds a master’s degree in cybersecurity and is CompTIA Security+ certified. His main interests are blue team operations, SOC workflows, threat detection, and practical ways defenders can use AI in their day-to-day work. He enjoys solving real-world technical problems, building his skills through hands-on labs, and sharing what he learns with others in the security community.

SpeakerBio:  Neha Gautam

Neha Gautam is a Product Manager at Microsoft Security and a Carnegie Mellon graduate specializing in the intersection of Identity and AI. From scaling platforms at Walmart to securing agentic AI at Microsoft, Neha’s mission is to translate complex security hurdles into seamless product experiences. She is a passionate mentor for women in tech and a frequent volunteer for WiCyS and Defcon Blue Team Village. Neha believes that the future of computing must be built by diverse perspectives—a philosophy she champions whether she’s designing enterprise governance or coaching early-career professionals.

SpeakerBio:  Omenscan ~

Some people write books to document and share what they learn. I write software. @Blueteamvillage Director. I do not speak for my employer, their clients, or customers

Creator: Bug Bounty Village

Many of us have heard that AI is just hype. But is that really the case? In this talk, Pedro “drop” Paniago walks through the experiment he conducted by applying AI to his bug bounty methodology, resulting in findings worth $50K in just a few weeks. Through real examples, lessons learned, current limits, and practical tips, he explores the impact AI is already having on bug bounty and security research.

People:
    SpeakerBio:  Pedro “drop” Paniago

Pedro Paniago also known as “drop” in the bug bounty community, is an Offensive Security Manager specializing in Application Security at PwC, as well as a security researcher with a strong focus on AI security. As a bug bounty hunter, he has identified more than 1,000 vulnerabilities and holds 10+ CVEs. He is certified in CBBH, eJPTv2, PJMT, and BSCP. In 2024, Pedro ranked in the top 3 at Belgium’s Hack the Government Live Hacking Event, and in 2025 he placed in the top 6. He is an active voice in the bug bounty community, a HackerOne Brand Ambassador, and co-captain of the Belgian team.

Creator: OWASP Foundation

Most AI security assessments test the chatbot and miss the system. The founding technical lead of the OWASP GenAI Security Project covers what actually breaks when you pentest AI applications across three distinct attack tiers — model exploitation, application-layer flaws, and agentic architecture failures — drawing from frontier model red team operations, bug bounty submissions, and offensive cybersecurity evaluations.

People:
    SpeakerBio:  Ads Dawson, Staff AI Security Researcher, OWASP GenAI Security Project founder

Ads Dawson founded the OWASP GenAI Security Project and led it to flagship status — the fastest in OWASP history. As a Staff AI Security Researcher at Dreadnode, he specializes in exploiting ML and AI systems, harnessing AI for offensive cybersecurity through capability development and exploit development, and conducting red team operations against frontier models for government, military, and tier-1 labs. He is lead author of AIRTBench (arXiv), the first benchmark for autonomous AI red teaming in LLMs, and author of AI Native LLM Security (Packt, 2025).

A senior red team operator with BT6 (the frontier AI red team), ranked #2 in Canada on HackerOne (2025/2026), and selected for Meta’s MBBRC live hacking event, Ads is a HackerOne US South Ambassador, BugCrowd Hacker Advisory Board member, MITRE AI Working Group contributor, and leads the OWASP Toronto chapter. He spoke at Bug Bounty Village DC33 on the BT6 AI jailbreaking panel and is also presenting “Exfil Everything: A Year of Stealing Data from AI Agents” at Bug Bounty Village DC34 (schedule pending publication).

Creator: Policy @ DEF CON

New York’s RAISE Act is among the most significant state AI safety laws in the country. Signed December 19, 2025, it imposes transparency requirements, safety plans, 72-hour incident reporting, and annual audits on frontier AI developers. It took effect 8 days after a presidential executive order authorizing the DOJ to challenge state AI laws. This talk applies the security researcher’s methodology: map the surface, find the gaps, assess whether compliance cost is proportionate to actual risk reduction. The structural gaps are specific. The $500M revenue threshold excludes the most aggressive AI capability development — well-capitalized but pre-revenue. The 10²⁶ FLOPs definition may not capture dangerous fine-tuned derivatives of frontier models. The catastrophic harm threshold requires 100+ deaths before reporting obligations attach — the alarm fires after the building has already burned. And the law requires transparency, not prohibition: a fully compliant developer can keep operating a system causing significant sub-catastrophic harm. 12 or more states are expected to model AI legislation on New York and California in 2026. If the security community does not engage critically with what this law actually does, those copies will inherit its gaps.

People:
    SpeakerBio:  Joshua Marpet, Finite State

Joshua Marpet is Senior Product Security Consultant at Finite State, co-host of Paul’s Security Weekly and Security Weekly News, and a CMMC co-author. He holds membership on the IEEE/UL 2933 and SPDX standards committees and serves on the boards of BSidesDE, Skytalks, and the Value Chain Risk Institute. A former police officer, firefighter, and Federal Reserve Bank of Philadelphia security analyst, he works at the intersection of security standards, policy, and operational practice.

Creator: Blue Team Village

It starts with a simple prompt.

An AI assistant writes production code in seconds. A phishing email becomes nearly indistinguishable from a legitimate one. A security analyst uses an AI copilot to investigate hundreds of alerts—while an attacker uses the same technology to automate reconnaissance, create malware, and scale attacks.

Welcome to the new reality of cybersecurity.

AI is changing how we build, defend, attack, and trust technology. As organizations rapidly deploy AI applications and autonomous agents, defenders face two urgent questions: How can we use AI to defend better, and how do we secure AI itself?

Join AI and cybersecurity practitioners for an honest discussion that separates hype from reality. Through real-world stories and practical insights, we’ll explore how attackers are using AI, how defenders are responding, what it means to secure AI systems, and where the industry is headed next.

Whether you’re a seasoned blue teamer, security leader, AI enthusiast, or new to cybersecurity, you’ll leave with a clearer view of today’s AI security landscape—and what tomorrow’s defenders need to know.

The future of cybersecurity is already here. Are you ready?

People:
    SpeakerBio:  Karan Dwivedi

Karan Dwivedi is a recognized cybersecurity expert. Currently, he serves as a security engineering manager at Google. Karan has led large-scale security projects at Google and Yahoo in the US for products like Google Search, Google Assistant, Yahoo Mail, Yahoo Finance, Flickr, etc, to safeguard over a billion users. At Yahoo, he was part of the security team responding to the world’s largest data breach. He is the author of the book “Kickstart your security engineering career” which is a definitive guide for anyone looking to start a career in security engineering. Karan contributed to the latest internet standard for scoring vulnerabilities, the Common Vulnerability Scoring System (CVSS 4.0). He is featured in major media like Hakin9 Media Magazine, Forensic Focus News, etc. He has delivered talks at national and international conferences like Tech Ex North America, Tech Summit SF, BSides Las Vegas, National Cyber Summit, etc, to influence private and public sectors. Karan was featured as a subject matter expert in the Google Cybersecurity Certificate program launched in May 2023 on Coursera, which had an enrollment of over 41000 students in a few weeks. Furthermore, Karan has served as an advisor to startups, an editorial board member in international security journals, and judged global competitions. He holds a master’s degree in Information Security from Carnegie Mellon University, USA. His portfolio can be found at https://karand.me

SpeakerBio:  Thomas Roccia

Thomas Roccia is the Founder and Threat Researcher at SecurityBreak, a company dedicated to AI Threat Intelligence and AI security. With more than 15 years of experience in cybersecurity, he has investigated major cyberattacks, led threat research at Microsoft and McAfee, collaborated with law enforcement, and helped organizations understand and respond to emerging threats.

Speaker / Contributor: This individual participates in SANS events, presentations, written content, or other community resources as an external contributor. They are not a SANS employee, instructor, or affiliate.

Since 2022, Thomas has focused on AI Threat Intelligence, researching how attackers target AI systems and how defenders can secure the growing AI ecosystem. He is recognized as one of the early pioneers in applying generative AI to cyber threat intelligence through practical research, open-source projects, and industry education.

Thomas is the creator of several widely used open-source initiatives, including the Unprotect Project and NOVA, one of the first detection-rule frameworks for prompt pattern matching and AI security monitoring. NOVA won the SANS AI Cybersecurity Hackathon in 2025, and Thomas received the SANS Difference Makers Award (DMA) the same year for his contributions to the cybersecurity community. He is also the author of the bestselling book Visual Threat Intelligence: An Illustrated Guide for Threat Researchers, recipient of the Bronze Foreword INDIES Award in the Science & Technology category.

A regular speaker at leading security conferences, including BlackHat, DEFCON, BSides, and SANS events, Thomas teaches AI Threat Intelligence to security professionals around the world, to help defenders apply AI to real-world investigations while understanding the risks attackers pose to the AI ecosystem.

Through SecurityBreak, Thomas develops research, intelligence, and products that help organizations monitor, detect, and defend against attacks targeting AI models, agents, tools, and the broader AI ecosystem.

SpeakerBio:  Todd Fletcher

Todd Fletcher (@kobaltfox) has spent 25 years on the defense side, running IT and security for public government, wiring up SIEM and SOAR pipelines, and building security programs that survive contact with a real budget. He works at CrowdStrike as an AI Services Technical Lead in Strategic Advisory Services, where a lot of his job is separating what AI actually does for a SOC from what the slide deck claims it does.

He is also a PhD candidate in cyberpsychology at Birmingham City University (UK), studying the people doing the defending: what drives them, what burns them down, and why “humans are the weakest link” has always been a lazy answer. He writes at kobaltfox.com and toddmfletcher.com which starts from the position that defender mental health is a security outcome, not a wellness perk.

Creator: AI Village

Creator: AI Village

Creator: Telecom Village
Key Discussion Points:
1. AI vs. AI Cyber Warfare: How AI is transforming both cyber defense and cyber offense in telecom, and whether autonomous attacks are already becoming a reality.
2. AI-Powered Telecom Security: The role of AI in SOC operations, fraud detection, threat intelligence, network security, and automated incident response.
3. Securing AI & AI Governance: Protecting AI models from threats such as prompt injection, data poisoning, model theft, and ensuring secure AI adoption through governance and human oversight.
4. Future of Telecom Security: AI-native networks, OEM and supply chain security, the evolution of security professionals, and preparing telecom operators for the AI-driven threat landscape by 2030.

People:
    SpeakerBio:  Rohini
No BIO available
SpeakerBio:  Pankaj Sontakke
No BIO available
SpeakerBio:  Will Thomas
No BIO available
SpeakerBio:  Isabel Manjarrez, Threat Researcher

[EN] María Isabel Manjarrez is a security researcher with Kaspersky’s Global Research and Analysis Team (GReAT). She specializes in investigating threat actors in Latin America, tracking their movements, and analyzing the new techniques they deploy. She holds a degree in telecommunications and electronic systems engineering and her interests include threat intelligence, malware analysis, satellite communications, electronics, and music.

She has shared her knowledge as a speaker at local and international conferences such as Defcon, Security Analyst Summit (SAS), and EkoParty.

---

[ES] María Isabel Manjarrez es investigadora de seguridad en el Equipo Global de Investigación y Análisis (GReAT) de Kaspersky. Se especializa en la investigación de actores amenaza en Latinoamérica, rastrea sus movimientos y analiza las nuevas técnicas que implementan. Es Ingeniera en telecomunicaciones y sistemas electrónicos, sus intereses incluyen la inteligencia de amenazas, análisis de malware, comunicaciones satelitales, electrónica y música.

Ha compartido su conocimiento como ponente en conferencias locales e internacionales como Defcon, Security Analyst Summit (SAS) y EkoParty.

SpeakerBio:  Omer Farooq

Omer Farooq is a cybersecurity, cloud, and artificial intelligence leader with more than twenty-five years of experience spanning application security, cloud architecture, software engineering, DevSecOps, AI security, and technology innovation. As Founder and Principal Security Consultant at Auxin Security, Omer has advised more than 100 organizations worldwide, including Fortune 500 companies, government agencies, healthcare organizations, and nonprofits. His expertise includes GenAI and LLM security, offensive security assessments, threat modeling, cloud security, DevSecOps transformation, secure software development, and enterprise architecture. Omer is a frequent speaker at industry conferences and events including RSA Conference, BSides DC, AWS events, NAB Show, Microsoft Azure conferences, and numerous cybersecurity forums. His current research focuses on AI security, agentic systems, retrieval-augmented generation security, offensive AI testing, and emerging threats targeting enterprise AI deployments.

SpeakerBio:  T -Mobile CTF Team
No BIO available
SpeakerBio:  James(GSMA)
No BIO available
SpeakerBio:  Arvind Singh
No BIO available

Creator: Radio Frequency Village

Software defined radios have lowered the barrier to RF experimentation, but GNU Radio still requires a working knowledge of blocks, port types, sample rates, hardware drivers, and connection rules. This talk presents an AI-assisted GNU Radio design system that lets a user describe an RF task in natural language, then uses a large language model to build, validate, and organize the resulting flowgraph through structured tool calls.

Direct LLM-generated GNU Radio code often looks plausible but fails in practice. Blocks may be incompatible, parameters may be invalid, connections may not match, and visual flowgraphs may be arranged in ways that make them difficult to inspect or modify. This system addresses those problems by exposing GNU Radio through validated tools for block discovery, hardware detection, flowgraph creation, parameter configuration, connection management, layout, and code generation. Before producing the final design, it checks port types, stream and message domains, vector lengths, and incompatible signal paths. When blocks cannot be connected directly, it recommends the required conversion blocks instead of generating a broken flowgraph.

This presentation will show the system live through practical SDR and RF security workflows. The goal is to make GNU Radio flowgraph development faster to use when exploring wireless systems, validating assumptions, and moving from an RF idea to a working flowgraph.

Links:
    Github – https://github.com/Dollarhyde/gr-mcp

People:
    SpeakerBio:  Erwin Karincic (Dollarhyde)

Erwin is an experienced security researcher specializing in both hardware and software reverse engineering, binary analysis, and exploit development across a range of processor architectures. He has notable experience in implementing complex Radio Frequency (RF) waveforms using Software Defined Radios (SDRs) for cybersecurity applications, complemented by his proficiency in designing, simulating, and fabricating antennas tailored for such applications. His past work includes extensive TCP/IP networking experience, designing worldwide secure communication systems. Erwin holds a number of prestigious certifications, including OSCP, OSCE, OSWE, OSEE, and CCIE Enterprise Infrastructure.

Creator: La Villa Community

¿Qué sucede cuando la barrera técnica para ejecutar ataques Living off the Land (LotL) desaparece y una IA toma el control?

Esta charla presenta AImaru C2, un framework ofensivo que expone el futuro del cibercrimen. Aimaru introduce varios conceptos nuevos para la industria como la utilización del Model Context Protocol (MCP) para transformar clientes inofensivos en Troyanos de Acceso Remoto (RATs). Estos agentes usan binarios nativos del sistema para operar de forma encubierta, orquestados íntegramente por un LLM sin restricciones mediante lenguaje natural.

Demostraremos cómo AImaru automatiza el cyberkill chain: desde bypasses dinámicos de AMSI generados al vuelo, jerarquización de tareas para Lotl, hasta la explotación de un vectores inéditos intenando abusar de IDEs locales (como Claude Code o PyCharm) para ejecución silenciosa.

Analizaremos cómo esto revolucionará el ecosistema RaaS/MaaS y los severos desafíos que enfrentarán los equipos de respuesta ante esta nueva generación de amenazas autónomas.

People:
    SpeakerBio:  Mario Lobo, Cyber Threat Intelligence Researcher

I am a Colombian Cybersecurity Engineer with a Master’s degree and a deep passion for the field. I have spent over 18 years immersed in various domains of Information Security and Cyber Intelligence.

My career has spanned critical sectors, including National Defense, where I spent nearly 10 years collaborating with Strategic Cyber Intelligence teams worldwide. I have led multiple cybersecurity teams for multinational banking institutions within the financial sector, and for several years now, I have served as the Lead Threat Intelligence Researcher at Lumu Technologies. I have been a main track speaker at prestigious international conferences such as Ekoparty, 8.8 Gob, BSides São Paulo, BSides Colombia, and Cyberwings, among others.

Beyond the professional realm, I am a guitar enthusiast, a dedicated cyclist, and a consummate music lover.

—

Soy colombiano, Ingeniero Magister en Ciberseguridad y un apasionado por el tema. Llevo más de 18 años sumergido diferentes campos de la Seguridad de la Información y la Ciberinteligencia.

Mi trayectoria me ha permitido transitar por sectores como la Defensa Nacional, donde por casi 10 años colaboré con equipos de Ciberinteligencia Estratégica alrededor del mundo. Lideré diferentes equipos de ciberseguridad para la banca multinacional en el sector Financiero y desde hace algunos años me desempeño como Lider Investigador de Inteligencia de Amenazas en Lumu Technologies. He participado como conferencista main track en eventos como Ekoparty, 8.8 Gob, Bsides Sao Paulo y Bsides Colombia, Cyberwings entre otros.

Fuera del ámbito profesional, soy un amante de la guitarra, la bici y un melómano consumado.

Creator: Red Team Village
AImaru C2 is the evolution of a concept sparked by an unsettling question: What happens when an attacker can execute advanced Living off the Land (LotL) attacks without being an expert? While traditional LotL requires a specialist to navigate complex environments, AImaru C2 changes the game. Evolving from a PoC into a sophisticated Red Team framework, it automates expert-level tradecraft by subverting Anthropic’s Model Context Protocol (MCP). By weaponizing MCP clients, it transforms a standard context protocol into a functional, autonomous multi-level RAT that natively “speaks” PowerShell.

AImaru C2 was born out of a critical necessity to ground cybersecurity in reality. While the industry often distracts itself with over-engineered narratives and hype-driven threats that ignore actual ecosystem data, a far more pragmatic danger has been hiding in plain sight. This framework addresses the unsettling evolution of the “Expert Bottleneck”: What happens when the sophisticated LotL tactics, once reserved for elite nation-state actors, are democratized through automation? Named after the mythical serpent that glides unseen between the underworld and the land of the living, AImaru C2 represents the shift from manual expertise to autonomous execution. It is no longer just a proof of concept; it is a full-scale Red Team framework designed to expose how easily an unskilled attacker can now weaponize authorized system protocols to orchestrate advanced, data-driven breaches that traditional defenses simply aren’t looking for.

To understand its impact, context is key. For years, LotL attacks have been the dominant tactic in ransomware campaigns. Instead of deploying traditional malware, threat actors abuse legitimate OS tools—such as PowerShell, WMI, or Certutil—to operate under the radar. In 2025, 82% of successful breaches were fileless or malwareless, with PowerShell ranking as the second most utilized TTP by groups like Black Basta, Royal, and LockBit.

This robust framework moves beyond experimental concepts, consolidating itself as a production-ready offensive platform. AImaru C2 is engineered with a modular architecture that prioritizes cryptographic integrity, operational security, and intelligent automation.

Core Framework Capabilities: – Model Context Protocol (MCP) RAT Engine: A paradigm shift in command and control that subverts Anthropic’s MCP from a standard context-sharing tool into a functional, undetectable Remote Access Trojan (RAT), tunneling all C2 traffic through legitimate LLM API communication. – PowerShell Client Builder: An automated obfuscation engine featuring deep variable randomization and structural mutation to break static signatures. – AMSI Bypass Generator: Dynamic generation of multi-stage bypass scripts, utilizing diverse techniques to neutralize the Antimalware Scan Interface in real-time. – Real-time Monitoring: Advanced telemetry for live command execution tracking and granular session management. – Cryptographic Isolation: Implementation of per-client encryption keys derived via HKDF-SHA256, ensuring that the compromise of one beacon does not jeopardize the entire fleet. – RBAC: Native Role-Based Access Control, strictly segregating permissions between Admins, Users, and Viewers. – Complete Audit Logging: Forensic-grade command history with millisecond-accurate timestamps for post-operation deconfliction.

By utilizing a multi-tier decision logic, the system dynamically selects and rewrites payloads based on the specific operational objective. This allows the C2 to scale its complexity in real-time—transitioning from basic PowerShell structures and stealthy WMI manipulation to the strategic abuse of LOLBins for high-complexity tasks. This adaptive approach ensures that every command utilizes the most effective legitimate system tool, successfully evading environment-specific defenses by blending into the target’s unique operational noise.

Tactic Description: In this interactive session, attendees will sit down with the developer to operate the AImaru C2 framework in a controlled lab environment. Participants will experience the “Natural Language Intent” paradigm shift firsthand, moving away from manual syntax to intent-based exploitation.

Deployment: Deploy a lightweight MCP-Client on a target Windows machine and establish a beacon back to the AI-C2 controller.

Intent-Based Recon: Issue high-level natural language prompts (e.g., “Find sensitive PDF files and identify lateral movement paths”) and observe the LLM selecting and generating the appropriate PowerShell/WMI code in real-time.

Automated Evasion: Trigger the Adaptive AMSI Bypass module, witnessing how the framework re-writes its own memory-resident payloads to evade active defenses.

All the technical information about the project is here: https://github.com/mloborom/aimaru-c2

People:
    SpeakerBio:  Mario Lobo, Cyber Threat Intelligence Researcher

I am a Colombian Cybersecurity Engineer with a Master’s degree and a deep passion for the field. I have spent over 18 years immersed in various domains of Information Security and Cyber Intelligence.

My career has spanned critical sectors, including National Defense, where I spent nearly 10 years collaborating with Strategic Cyber Intelligence teams worldwide. I have led multiple cybersecurity teams for multinational banking institutions within the financial sector, and for several years now, I have served as the Lead Threat Intelligence Researcher at Lumu Technologies. I have been a main track speaker at prestigious international conferences such as Ekoparty, 8.8 Gob, BSides São Paulo, BSides Colombia, and Cyberwings, among others.

Beyond the professional realm, I am a guitar enthusiast, a dedicated cyclist, and a consummate music lover.

—

Soy colombiano, Ingeniero Magister en Ciberseguridad y un apasionado por el tema. Llevo más de 18 años sumergido diferentes campos de la Seguridad de la Información y la Ciberinteligencia.

Mi trayectoria me ha permitido transitar por sectores como la Defensa Nacional, donde por casi 10 años colaboré con equipos de Ciberinteligencia Estratégica alrededor del mundo. Lideré diferentes equipos de ciberseguridad para la banca multinacional en el sector Financiero y desde hace algunos años me desempeño como Lider Investigador de Inteligencia de Amenazas en Lumu Technologies. He participado como conferencista main track en eventos como Ekoparty, 8.8 Gob, Bsides Sao Paulo y Bsides Colombia, Cyberwings entre otros.

Fuera del ámbito profesional, soy un amante de la guitarra, la bici y un melómano consumado.

Creator: Biohacking Village

A frugal, eDNA sniffing system to audit atmospheric biodata-leaks in the era of Climate change.

Current healthcare cybersecurity focuses on device access, ignoring the fundamental vulnerability: the environmental air gap is closing. Massive cryospheric melting is releasing a backlog of ancestral genetic data (eDNA) into the biosphere, creating an unmonitored ‘input stream’ of allergens and pathogens. This talk presents a resilient, open-source eDNA monitoring system designed to audit these atmospheric ‘data leaks’ in real-time. By utilizing high-fidelity open-source and decentralized brewed polymerases and consumer electronics, we provide a solution for biological sovereignty that remains operational even when traditional digital infrastructure fails.

Here, we will present how to build global atmospheric monitoring networks for under $500 USD to intercept “paleo-organisms”—dormant bacteria and viruses released by melting poles—using shotgun metagenomics, 3D-printed parts, and microcontrollers. Essentially, he catches genetic ghosts in the air before they trigger the next global health crisis. This projects aims to tackle a sci-fi level threat: Facing the looming ecological danger of “polar amplification,” David is developing open-source hardware together with Biohackers at BioOlympia (Olympia Washington), to capture and sequence ancient or threatening environmental DNA (eDNA). His atmospheric biosensors combine high-precision 3D printing, fluid dynamics, and embedded systems to track clouds of paleo-biomass and frozen microorganisms waking up from the ice or perturbed ecosystems. Along with studies in environmental science, biology, and Geographic Information systems, BioOlympia also operates a fully equipped BSL-2 research environment.

People:
    SpeakerBio:  David J. Castillo-Cornejo

David is a biomedical scientist with experiences at the Max Planck Institute, Osaka University and ASU. He is the founder of Glyxon Biolabs. BioOlympia is a [Bio]Punk laboratory. The lab conducts advanced research in mammalian cell biology, CRISPR-based gene editing, and regenerative medicine.

Creator: Voting Village

The landscape supporting election cybersecurity has shifted dramatically since 2024. The federal government has lost most of its institutional knowledge about election security. Federal agencies have withdrawn from this space and diminished support for the Election Infrastructure Information Sharing and Analysis Center, leaving election officials and their vendors to shoulder the cybersecurity burden on their own. State and local election officials describe a breakdown in trust, reduced support, less intelligence, and other caustic effects that are debilitating the information-sharing environment. Meanwhile, advanced threats and threat actors have not gone away.

The fracturing of election security support goes beyond funding cuts for threat monitoring and fewer cyber assessments. Breakdowns in the multilateral information sharing mean that election officials lose early warnings, resource-strapped jurisdictions lose collective knowledge, and, critically, the federal government loses the capability to detect foreign campaigns to interfere in US elections.

An ineffective information-sharing environment leaves state and local jurisdictions facing sophisticated adversaries and criminal organizations without collective defense. Some losses can be replicated, but the gaps leave election security weaker than before.

Civil society, particularly former federal, state, and local election security professionals, is attempting to fill these gaps. Organizations such as the Election Security Exchange, Center for Democracy and Technology, Committee for Safe and Secure Elections, Partnership for Elections and Emergency Response, ASU Mechanics of Democracy Lab, MS-ISAC, Election Technology Education Fund, and other partners are rebuilding collective defenses. This panel acknowledges that the environment is what it is and demonstrates that vulnerabilities are not going unrecognized or unaddressed.

People:
    SpeakerBio:  Geoff Hale

Geoff Hale has over 15 years of experience supporting the security of critical infrastructure with a focus on election security since 2016. Geoff worked at the Cybersecurity and Infrastructure Security Agency (CISA) as Associate Director for Election Security & Resilience, helping to build a security community in support of U.S. elections.

SpeakerBio:  Ryan Macias

Ryan Macias has spent over 20 years providing subject-matter expertise in election technology, security, and administration to election officials across the U.S. and election management bodies (EMBs) abroad. Macias has advised thousands of election stakeholders on strategies and methods to build resilience in election infrastructure, technology, and processes. Macias also teaches Election Security at the University of Minnesota and previously worked for the U.S. Election Assistance Commission (EAC) as the Acting Director of the Voting Systems Program and California Secretary of State, where he led the Top-to-Bottom Review of Voting Systems.

Creator: DCNextGen

Learn about EFF’s work protecting your right to send memes to your friends, watch videos online, and go outside without surveillance. Ask questions about EFF’s work, from how we fight bad laws in Congress to how we take on big tech companies. Whether you’re curious about technology, want to know how to protect your privacy, or just want to know what a nonprofit org actually does, this is your chance to ask us anything.

People:
    SpeakerBio:  EFF
No BIO available

Creator: Mobile Hacking Community

The Java programming language is the backbone of almost all Android applications, yet its compilation process is often discarded. Behind the scenes, a complex series of bytecode transformations takes place, and tools such as ASM allow anyone to hook into the process and alter the flow to achieve better performance, obfuscation, compile-time code injection, … In this talk, we’ll explore how this tool works, how it can be used, and the deep pitfalls to look out for when performing Java bytecode manipulation.

People:
    SpeakerBio:  Ricardo Loura
No BIO available

Creator: Call Center Village

Free cable box, or Chinese-controlled botnet node? Darknet Diaries Ep. 172 exposed SuperBox as both. Part 3 breaks down how one sketchy streaming device turned a living room into a national security case.

People:
    SpeakerBio:  D3adA55
No BIO available

Creator: Malware Village
Format: Talk
Length: 20-40min (I’m flexible, I can adjust the content based on how long the time slot is)

Abstract: In September 2025, an obscure npm package shipped with a piece of JavaScript that should not have been there. A new beast had emerged among supply chain attacks. Within 48 hours, that JavaScript had copied itself into more than 180 packages. It did something that hadn’t been seen before in such an attack. It was self-propagating. It was a worm. And it was LOUD. Hundreds of public repos were created, littered with stolen secrets. Within nine days, it had triggered an emergency CISA alert.

Two months later, we witnessed the “Second Coming”: a far more aggressive successor returned, racing a deadline npm itself had set, and leaving behind roughly 700 compromised packages, ~25,000 GitHub repositories of stolen secrets, and a destructive payload that wiped victims’ home directories when it couldn’t exfiltrate them. And these were just the first two attacks.

This talk is a technical breakdown of the Shai-Hulud worm family: the first malware to spread successfully through the npm ecosystem as a self-replicating worm, and the iterations that followed. We’ll examine how the malware infected its victims, turned legitimate security tooling against them, what it stole, and how it kept spreading.

Each iteration evolved to defeat the defenses put in place against the last, swapping tools, tactics, and even its JavaScript runtime. Stranger still: the malware wasn’t trying to hide. It branded its own exfiltration repositories, named its propagation functions clearly in source, and used victims’ own GitHub accounts to share stolen secrets. By the end, we’ll have a clearer picture of how these supply chain worms work, and what each rising of Shai-Hulud has taught us.

People:
    SpeakerBio:  Megg Sage, AppSec engineer who explains scary things about your dependencies

Megg is an application security engineer who started out as a web developer. Security drew her in with the endless puzzles and challenges put forth by the field. She loves sharing knowledge, particularly when she can both educate and frighten her audience at the same time. After all, what can happen when security goes wrong is pretty scary. She also enjoys working closely with software engineering teams to try to make security work within existing development practices or at least minimize the pain of “doing security.” When not behind a computer, Megg can usually be found making some sort of costume piece or shiny object.

Creator: Voting Village

The last time I spoke at DEF CON’s Voting Village, Curling v. Raffensperger was still awaiting a decision. Since then, the technical record has expanded, the litigation has largely run its course, and yet the strategic debate has changed remarkably little. High-consequence systems operate under two competing optimizations. Engineers optimize for reality. Administrators optimize for operations. Successful missions answer operational questions; they do not validate engineering assumptions. Yet documented engineering anomalies that do not lead to immediate mission failure gradually become accepted as normal. Researchers are repeatedly asked to demonstrate that documented engineering anomalies will produce catastrophic failure, while institutions are seldom asked to explain why anomalies have become compatible with normal operations. This burden-shifting has shaped two decades of election security research while limiting its strategic impact. The election security community has accepted the wrong burden. To the extent that security researchers participate in defining the safety boundaries of electronic election infrastructure, their role has largely been to discover and document anomalies and provide expert testimony in proceedings aimed at optimizing normal election operations. Lost in that process is the implication that “normal operation” has a factual, engineering basis. In practice, normality has become an institutional designation that researchers seldom contest. This talk argues that changing the burden fundamentally changes what institutions are permitted to call normal. It proposes a new research agenda focused less on discovering the next anomaly than on preventing documented engineering anomalies from becoming institutionalized as accepted practice. The implications extend well beyond elections. The forthcoming book Sovereign Code argues that the same organizational dynamics increasingly govern all software-mediated public administration. The success of the next generation of election security research will be measured not by the anomalies it discovers, but by the anomalies institutions are no longer able to call normal.

People:
    SpeakerBio:  Richard DeMillo

Richard DeMillo holds the Charlotte B. and Roger C. Warren Chair in Computing at The Georgia Institute of Technology in Atlanta, Georgia. He is the former Dean of Georgia Tech’s College of Computing, founding Chair of the School of Cybersecurity and Privacy, and founder of the Center for 21st Century Universities. In industry, he was previously Vice President and Chief Technology Officer for Hewlett-Packard, and General Manager and Head of Computing Research for Bellcore/SAIC. In government, he directed the Computer and Computation Research Division at the National Science Foundation and the Software Test and Evaluation Project for the Office of the Secretary of Defense. He is the author of more than 100 scientific articles, books, and patents in cryptography, cybersecurity, and software engineering. He is a fellow of the American Association for the Advancement of Science, the Association for Computing Machinery, and the Lumina Foundation.

Creator: IoT Village

Two years ago I gave a popular talk at Defcon called “Anyone Can Hack IoT” and the message was simple, you don’t need expensive gear and I can show you how. All of that is true, but now it’s even easier and I want to show you why. Whether you saw the original talk or this is your first time thinking about IoT hacking, I’ll show you how AI has actually made it even easier. In this talk I’ll show you what’s actually useful and what not by walking through my real workflow that I’ve used to find multiple CVEs. I’ll demo Wairz, an open source tool I built that lets AI agents help reverse engineering firmware, along with some other hardware tools I’ve put together to give AI direct access to devices on my bench. I’ll cover what’s worth using AI for, what’s still better done by hand and how you can build your own AI assisted setup at home and hack your first device (or second or third or so on).

People:
    SpeakerBio:  Andrew Bellini
No BIO available

Creator: Voting Village

People:
    SpeakerBio:  Matt Blaze

Matt Blaze is the McDevitt chair in Computer Science and Law at Georgetown University, where he studies problems at the intersection of technology and public policy. Election systems and voting technology are central focuses of his research. He led teams as part of the California TTBR and Ohio EVEREST studies that each found deep and fundamental weaknesses in different election technologies used by those states and the rest of the US. He has testified on technical risks in elections before the US Congress and other bodies numerous times. Blaze is a co-founder of the Voting Village and president of the Election Integrity Foundation.

SpeakerBio:  Geoff Hale

Geoff Hale has over 15 years of experience supporting the security of critical infrastructure with a focus on election security since 2016. Geoff worked at the Cybersecurity and Infrastructure Security Agency (CISA) as Associate Director for Election Security & Resilience, helping to build a security community in support of U.S. elections.

SpeakerBio:  Michael Moore

Michael Moore is the CISO at Arizona Secretary of State’s Office. He leads efforts to strengthen election security through coordinated partnerships across federal, state, and local governments, alongside trusted private-sector partners. His work is grounded in protecting democratic processes and maintaining public confidence in elections. He focuses on the most pressing risks to election integrity: mis-, dis-, and malinformation (MDM), and the insider threats that can emerge from an increasingly polarized and misinformed environment. His approach centers on countering false narratives with verifiable truth while advancing a defense-in-depth strategy. This includes preventing attacks wherever possible, rapidly detecting anomalies, and ensuring resilient, transparent recovery when incidents occur.

Through organizational leadership and national collaboration, Michael has driven initiatives that enhance the security, resilience, and credibility of election systems. His work helps safeguard the voter experience and uphold trust in democratic institutions.

SpeakerBio:  Kendall Spencer

Kendall Spencer is an attorney specializing in emerging companies, technology transactions, and the legal issues shaping innovation. Since joining DEF CON’s Voting Village as a Georgetown Law student in 2019, he has worked alongside Matt Blaze, David Jefferson, and the Voting Village team at the intersection of election technology, cybersecurity, and democracy. Spencer serves on the Board of Directors of the Election Integrity Foundation and has advised state election officials on election security, post-election audits, and cybersecurity best practices. His work focuses on bridging the gap between technical research, public policy, and the law—helping policymakers, election officials, security researchers, and the public better understand the role election security plays in strengthening democratic institutions and public confidence in elections. A frequent DEF CON speaker, Spencer is passionate about fostering thoughtful, bipartisan conversations on the role of technology in protecting election integrity and the democratic principles that underpin a free and open society. Outside of his legal and technology work, Spencer is a rare book dealer and collector specializing in early American history and the Black diaspora.

SpeakerBio:  Philip Stark

Philip B. Stark is Distinguished Professor of the Graduate School at the University of California, Berkeley, where he has served as department chair and associate dean. In 2007 he invented “risk-limiting audits” (“RLAs”), endorsed by the National Academies of Science, Engineering, and Medicine and the American Statistical Association, among others, and required or authorized by law in about 15 states. He designed and helped conduct the first dozen pilot RLAs, helped draft RLA legislation for several states, and has published open-source software to support RLAs. In 2012, he and David Wagner introduced “evidence-based elections,” a paradigm for conducting demonstrably trustworthy elections. Stark has served on the Board of Advisors of the US Election Assistance Commission and its cybersecurity subcommittee, the Board of Directors of Verified Voting Foundation and the Election Integrity Foundation, and on the California Post Election Audit Standards Working Group. He has worked with the Secretaries of State of California, Colorado, and New Hampshire and numerous local election officials. He has testified about election integrity in state and federal courts and to legislators. He received the IEEE Cybersecurity Award for Practice, the UC Berkeley Chancellor’s Award for Research in the Public Interest, and the John Gideon Award for Election Integrity. He is a fellow of the American Academy of Arts and Sciences, the American Statistical Association, and the Institute of Physics.

Creator: Policy @ DEF CON

An energy provider detects an intrusion when a control-room operator notices an unexplained setpoint change not an alert. Initial access occurred ninety days earlier via a contractor credential that outlived contract. Disclosure clocks are already running. Scope is unknown. Systems cannot go offline. This reflects recurring patterns Microsoft incident response teams observe across critical infrastructure, patterns policy was not designed around today. Cyber policy assumes breaches are detectable in reasonable timeframes, attributable with confidence, and containable once identified. Incident response shows otherwise. PRC state-sponsored actors such as Volt Typhoon maintain persistent, low-visibility access using valid credentials and living-off-the-land techniques. Detection often lags weeks or months, triggered by operational impact rather than telemetry. Investigations unfold under live operations, incomplete visibility, and uncertain scope, while disclosure timelines under CIRCIA, SEC rules, NIS2, and similar frameworks continue to run. This session bridges incident response and policy design, drawing on CI cases, DOJ experience, crisis counsel to show where assumptions break and how policy can align to conditions: identity-driven intrusion, persistent access, IT-OT convergence, and decisions under uncertainty.

People:
    SpeakerBio:  Travis Berent, Microsoft

Travis Berent is the Director of Security Business Strategy at Microsoft, where he helps shape the company’s approach to incident response, resilience, and cybersecurity partnerships. Prior to joining Microsoft, Travis served as Director for Cybersecurity Policy and Incident Response at the National Security Council, where he strengthened U.S. defense and intelligence environments, oversaw response to significant cyber incidents affecting critical infrastructure and government networks, and led key processes including the Vulnerabilities Equities Process. He also helped develop national policy responses to ransomware and contributed to both the National Security Strategy and National Cyber Strategy. Earlier in his career, Travis served in multiple roles at the Federal Bureau of Investigation (FBI), including in the New York Field Office and at Cyber Division Headquarters, where he worked on counterintelligence and cyber matters.

SpeakerBio:  Adam Hickey, Mayer Brown

Adam Hickey offers clients extensive experience in cybersecurity and national security matters. Adam draws on more than 15 years of experience at the U.S. Department of Justice (DOJ) handling high-profile national security matters intersecting with the private sector. Prior to joining Mayer Brown, he was a Deputy Assistant Attorney General (DAAG) of DOJ’s National Security Division (NSD). During his time at NSD, Adam established DOJ’s national security cyber program, dedicated to combatting malicious cyber activity by foreign intelligence services affecting major companies and critical infrastructure, and he supervised the criminal investigation and charging of every such case for more than a decade. He now advises critical infrastructure and OT operators on incident response, privileged investigations, disclosure, regulatory requirements across borders, and enforcement actions.

Creator: Red Team Village

This VR visualization experience is designed to support the exploration and understanding of complex cybersecurity scenarios for students and professionals. Cyber attack campaigns often involve intricate relationships and multi-stage processes that can be difficult to interpret, making effective analysis and remediation challenging. By leveraging visualization best practices in virtual reality, the experience presents cyber threat information in a more intuitive and interactive format, with the goal of improving comprehension, situational awareness, and the ability to identify meaningful and actionable insights.

People:
    SpeakerBio:  James Rice

Mr. James Rice has been cybersecurity faculty for the last decade in Upstate New York at Mohawk Valley Community College and more recently Rochester Institute of Technology. During this time, Mr. Rice has focused on developing numerous interactive gamified learning scenarios for the classroom and cyber competitions such as the NSA sponsored NCAE Cyber Games. Mr. Rice is currently pursuing his PhD at RIT in Computer Engineering and researching how to best leverage immersive reality technologies for data visualization and interaction, primarily in cyberspace.

Creator: Bug Bounty Village

Finding one prompt injection isn’t the end goal; checking how many variants still bypass your safeguarding layers is the new target. Introducing MPG, an XPIA mutation framework that generates adversarial payload variants and tests them across agentic AI systems to expose hidden vulnerability surfaces and emerging data exfiltration risks.

People:
    SpeakerBio:  Bogdan Stelee, Senior Software Engineer, Microsoft

Bogdan Stelea is a Senior Software Engineer performing AI Security Research at Microsoft, working at the intersection of AI security, adversarial ML, and secure agentic system design. Bogdan develops automated frameworks for conducting variant hunting and discovering prompt injection vulnerabilities at scale. His work focuses on identifying, reproducing, and understanding vulnerabilities in modern AI copilot solutions, with a particular emphasis on cross/indirect prompt injection attacks (XPIA), tool misuse, and the expanding attack surface introduced by AI agents. Moreover, Bogdan creates learning materials and courses that clearly explain the threats introduced by XPIA. Actively engaged in the AI Security, CTF, and Bug Bounty communities, with a consistent presence at DEF CON, Bogdan has published and presented his AI security research at international academic venues, including IEEE and ACM.

SpeakerBio:  Arnav Garg, Security Researcher, Microsoft

Arnav Garg is a Security Researcher at the Microsoft Security Response Center (MSRC), specializing in AI security research, adversarial testing, and the security of Copilot and agentic AI systems. His work focuses on understanding and mitigating emerging AI threats, with particular emphasis on cross/indirect prompt injection attacks (XPIA), data exfiltration techniques, tool misuse, and the evolving attack surface introduced by AI agents. Arnav develops automated frameworks for scalable security testing and vulnerability discovery, helping advance practical defenses for next-generation AI systems. He has presented and published research on AI security, automated prompt injection discovery, and AI vulnerability testing at both internal and external venues.

Creator: Red Team Village

Deploying a C2 server manually is a time consuming, sometimes annoying process. Deploying a C2 server weekly for two months, even more so. This Tactic will cover Gaia, the tool developed by the Midwest Collegiate Cyber Defense Competition Red Team to handle automated C2 deployment, user provisioning, payload creation, and more.

This Tactic will have attendees start with a standard Debian VM, and end with a fully provisioned Mythic server that can serve as the foundation for your next CTF, lab, or with a little extra work, your next op.

People:
    SpeakerBio:  Shad Brown

Shad is an Associate Adversary Simulation Consultant at SpecterOps where he performs a variety of offensive security assessments. Prior to his time at SpecterOps, he had roles in vulnerability management and network security engineering.

He enjoys teaching and mentorship and has spoken to students and faculty at dozens of high schools, colleges, and universities across the United States. In the last few years, he has volunteered as the red team lead for the states of Minnesota and Indiana in the Collegiate Cyber Defense Competition.

Creator: Cloud Village

Creator: La Villa Community

Autonomous, LLM-powered vulnerability scanners now crawl the web at scale — and every one of them shares the same structural weakness: it reads the content you control. This talk flips the engagement. Instead of just blocking the bot, your site detects that an AI agent is scanning it, fingerprints the model behind it, and fires back a hidden, tailored prompt injection that derails, poisons, or neutralizes the attacking agent — with no traditional exploit involved. We build on recent academic work on prompt-injection “hack back,” push it into the reconnaissance-and-scanning threat model with live demos, walk through an escalating cat-and-mouse between attacker and defender, and take an honest look at where active defense crosses the legal line. Red teamers and blue teamers alike will leave with something they can responsibly test the same week.

People:
    SpeakerBio:  Alcyon Junior, Head of Offensive Security & Fraud Hunting at Apura

Alcyon Junior is the Head of Offensive Security & Fraud Hunting at Apura, a speaker, writer, EXIN-accredited instructor, SecurityCast (securitycast.com.br) podcaster, and professor. He holds three degrees in Information Technology with a focus on computer networks, is a Cisco-certified specialist in Computer Networks, and has an MBA in IT Governance. Alcyon earned a Master’s degree in Knowledge Management and Information Technology with a focus on Cybersecurity from the Catholic University of Brasília, and he is currently pursuing a PhD in Electrical Engineering with an emphasis on Information Security and Communications at the University of Brasília (UnB). His international certifications include CompTIA Security+, Ethical Hacking Foundation (EHF), ISO/IEC 27002 (ISFS), ITIL® Foundation, Cisco Networking Academy (CNAP), McAfee Vulnerability Manager (MVM), and Server Professional (LPIC-1)

Creator: Cloud Village

CloudBashing started with reversing the private AWS, Azure, and GCP CloudShell REST protocols, analyzing websocket terminal sessions, and tracing janky browser authentication/credential flows with cookies and OAuth tokens. Along the way, we automated the APIs to access free CPU/networking, maintained access across container/VM resets, utilized persistent $HOME for implants/data, locked users out of sudo access, and installed a C2 framework. We discovered IAM design issues like: AWS role assumption that result in a large # of environments per compromised identity, web socket sessions that survive API token revocation, and M365/Gmail consumer email accounts that have default CloudShell access. This turned into a newly released exploit toolkit, CloudBasher, that enumerates, validates, installs, and runs distributed workloads with virtual storage and private networking across a large-scale agent network with persistence and resilience. We’ll demo distributing CPU-intensive workloads, using virtual storage for staging/exfiltration, secure networking for proxy and obfuscated exfil paths, while automating the discovery, enumeration, creation of CloudShell environments from initial credentials/sessions to implants/setup/networking to management and control.

People:
    SpeakerBio:  Jenko “edleft” Hwong, Huntress Labs

Jenko Hwong is a Principal Security Researcher at Huntress Labs, focusing on identity-based attacks and cloud abuse. Prior to Huntress, he spent 6 years at Netskope Threat Labs, has spoken at RSA and DEFCON, and is a Cloud Village Lead. He has over 20 years at various security startups in cloud detection/response, vulnerability scanning, AV/AS, pen-testing/exploits, L3/4 appliances, threat intel, and windows security.

SpeakerBio:  Chris Ryan, Huntress Labs

A series of oddly configured server banners, a JARM fingerprint, curious fields in a security certificate – these aren’t just technical details, but are instead threads in a narrative tapestry woven like a John le Carre novel. For over 20 years, Chris has dedicated his life to studying these threads and the intersection between cybersecurity, Russian linguistics, and free and open source software. His career path has taken detours through academia, aerospace and defense, software development, and cybersecurity.

Creator: Noob Community

A career journey may have many twists and turns – mine certainly has! This talk will explore some themes from my pivot to cybersecurity and my journey through threat intelligence, a offensive security startup, and a Federally Funded Research and Development Center (FFRDC). We’ll explore ways to advocate for yourself, suggestions for pivots and early career choices, and wrap it all up with thoughts on what can be next. Parts of the talk will be interactive and questions are encouraged.

So, what do you want to be and how can agency and being your own agent help you in your career journey?

People:
    SpeakerBio:  Meghan Jacquot
No BIO available

Creator: Red Team Village

In penetration testing, we constantly find ourselves facing EDRs/AVs, etc. But as in all cases, we aim to develop our own tooling or techniques that we believe may not yet be properly mapped or detected. On this occasion, we share part of our perspective when dealing with EDRs using Machine Learning technology configured in “FULL” mode. We were able to leverage this technique combined with an encrypted payload to evade this configuration; additionally, when the payload was downloaded or shared as a “.zip,” the agent did not inspect it. Now, let’s dive a bit into the technical aspects, but especially into how it was exploited. HookChain is an advanced technique for evading EDR solutions. It is primarily used by threat actors (APTs), but it can also be applied ethically in Red Team exercises. Its main principle is to intercept the hooks that EDRs install in the operating system kernel. What is a hook? Hooks are a mechanism used by EDRs to monitor calls to Windows API functions, mainly from ntdll.dll. In this way, they can intercept and analyze suspicious behavior in real time. The HookChain technique aims to bypass this monitoring layer.

Encrypting our payload with XOR HookChain typically combines hook evasion with payload encryption; in this case, we will use XOR to avoid static signature-based detection. The shellcode is encrypted in memory and only decrypted right before execution.

We can use a string (in this case, “CHANGEMYKEY”) as the key for our XOR encryption, along with shellcode generated using msfvenom, although we can also leverage other C2 frameworks, as we will see later.

People:
    SpeakerBio:  Arnold Jared Morales Yepez, Senior Team Lead, Grupo Salinas

Self-taught in computer security since age 12; holds a degree in Computer Forensics and Cybersecurity and is pursuing a Master’s in AI and Cybersecurity.

—

Desde los 12 años, me he dedicado a la informática con un enfoque especial en la seguridad. Actualmente, a mis 22 años, sigo explorando este mundo con la misma pasión, impulsado por la constante evolución de la tecnología y su interminable curva de aprendizaje.

Cuento con una carrera en Cómputo Forense y Ciberseguridad, actualmente curso una maestría en Inteligencia Artificial y Ciberseguridad.

Certificaciones: CWEE | CAPE |CPTS | EWPTX | CRTO | eMAPT | OSCP | OSCP+ |CEH V13 | EJPT| HTB prolabs Hades – Cybernetics – Zephyr – APTlabs | MDK

Creator: Mobile Hacking Community

About a year ago, Hakai Security introduced the Beerus Framework, a proposal to unify mobile tools in order to streamline the penetration testing process directly on Android devices. The idea was simple: centralize tooling and optimize the testing environment setup, removing the need for complex configurations and reducing operational friction during assessments.

Today, after a year of evolution, the framework is no longer just a concept. It has become a robust platform for offensive mobile analysis, gaining international recognition within the community. In this talk, we revisit Beerus in its current form, exploring its evolution, the capabilities that have been consolidated over time, and the lessons learned from real-world scenarios.

We also discuss the future vision of the project, covering new features, technical challenges faced during its development, and how the framework can continue evolving to keep up with modern mobile security assessment demands.

Official blog post: https://hakaisecurity.io/beerus-framework-a-new-mobile-framework-arises/insights-blog/ Repository: https://github.com/hakaioffsec/beerus-android Community references: – mobile hacker (post): https://x.com/androidmalware2/status/1979100030884618375 – mobile hacker (video): https://www.youtube.com/watch?v=8bDQzqw0_Sc – ZeroDay Gym (video): https://www.youtube.com/watch?v=I81xuVDsytE

People:
    SpeakerBio:  João Pedro Tricta, Client-Side Applications Squad Leader, Researcher, and Malware Developer at Hakai Offensive Security

Known as Tricta, I am 20 years old, Brazilian, and work as a Client-Side Applications Squad Leader, Researcher, and Malware Developer at Hakai Offensive Security. I am passionate about Sysinternals, reverse engineering, low-level, and client-side applications. In my free time, I enjoy programming, gaming, and watching anime. I’m a cat lover and a compulsive pizza eater.

SpeakerBio:  Daniel “Daniboy” França Lima, Pentester / Researcher at Hakai Offensive Security

Known as Daniboy, I am 19 years old, Brazilian, and work as a Pentester and Researcher at Hakai Offensive Security. I am passionate about understanding how things work, building tools, and manipulating systems. In my free time, I enjoy gaming, especially titles like Hollow Knight. I also love foxes, they are just too cute :3

Creator: Aerospace Village

The Rare Circuits team decided this was the year to take all our hard work on last year’s Aerospace Village badge and open source the useful components for the Badge Maker Community. We took the tight communication integration between badge and Add-on to create SAOv3, a new SAO standard that puts connectivity right at the forefront. ANSM, our framework to create silky smooth pixel animations and interfaces for embedded systems, is ready for everyone to use. We are even releasing our circuit and driver to make the SSD1106 OLED update smoothly and enables 4-level greyscale on it.

People:
    SpeakerBio:  Adam Batori
No BIO available
SpeakerBio:  Kevin Colley
No BIO available
SpeakerBio:  Robert Pafford
No BIO available
SpeakerBio:  Lillian Ash Baker
No BIO available

Creator: Nix Vegas Community

Home Assistant ships a warning that your install “will not work after 2026.11.” DuckStation detects NixOS four different ways and fatally errors your build on purpose. A maintainer won’t merge the one-line word-wrap fix you’ve needed for a year. On most platforms your options are comply or quit. On Nix, everything between you and the source is negotiable.

This talk is a guided tour up the override ladder, from module options and mkForce to overrideAttrs, fetchpatch-ing unmerged PRs, overlays, and finally patching nixpkgs itself. The stories along the way are real and primary-sourced: sabotaged npm packages, the xz backdoor, the youtube-dl DMCA takedown, rug-pull relicensing, and the nixpkgs patch literally named nixos-was-never-supported.patch.

People:
    SpeakerBio:  Daniel Baker

Daniel Baker is a developer, engineer, and mathematician passionate about reproducible software. An active NixOS community member and educator, he authored the NixCon NA 2024 Nix module system workshop and nixos-modules-lessons. He serves on the NixOS Marketing Team and helps organize both Nix Vegas and Planet Nix. He believes any system worth building is worth rebuilding, bit for bit.

Creator: Blacks In Cyber Village

In an industry that frequently highlights a massive skills gap, Black professionals are dramatically underrepresented in cybersecurity. This talk uses sharp humor and real-world experiences to explore the realities of being Black in the hacking community. It addresses challenges like the “only one in the room” tax, costly certifications, unconscious bias in hiring, and the emotional labor of code-switching. Attendees will gain insights into systemic barriers and learn how resilience and community, inspired by trailblazers like Camille Stewart Gloster and Tennisha Martin, can overcome rigged systems. This session also challenges allies and leaders to sponsor talent, improve hiring processes, and understand why increasing Black representation is essential for equity and innovation in cybersecurity.

People:
    SpeakerBio:  Dr. Louis DeWeaver III, Cybersecurity Consultant at Marsh McLennan Agency (MMA)

Dr. Louis DeWeaver is a battle-tested Cybersecurity Consultant at Marsh McLennan Agency (MMA) with over 20 years of real-world experience in the trenches. They hold a hard-earned academic arsenal: an Associate of Applied Science in Information Technology, a B.S. in Information Systems Security (2011), an M.S. in Information Assurance (2016), and a Doctor of Computer Science specializing in Cybersecurity (2021). Dr. DeWeaver doesn t just collect credentials they weaponize them. They serve on the MITRE Engenuity ATT&CK® Evaluations Community Advisory Board, helping define how organizations should actually defend against real adversaries. A proven competitor, they crushed the ONCD Badge Challenge at DEFCON 31 and 32, and in 2025 took down the Policy Village CTF badge created by former ONCD staff. Blunt, high-energy, and unapologetically direct, Dr. DeWeaver has delivered hard-hitting talks at Black Hat, DEF CON, GrrCon, and other major security events. They don t sugarcoat the industry s failures they call them out and show you how to survive (and win) anyway.

Creator: Red Team Village

This talk aims to demonstrate the modus operandi of an APT focused on full organizational compromise, revealing how such operations actually function in practice — not through isolated tools or techniques, but through the decision-making model that guides every action over time.

Rather than focusing on direct exploitation or individual methods, this session shows how advanced operators structure their actions around continuity, predictability, and behavioral alignment with the target environment. Attacks are not treated as disruptive events, but as controlled sequences of decisions designed to remain coherent and below the threshold of attention.

The presentation walks through a real RedOps engagement from a strategic perspective, highlighting how each step is evaluated in terms of detection probability, behavioral deviation, and operational impact, ensuring the operation remains stable and does not generate relevance over time.

Instead of teaching specific techniques, the talk focuses on analyzing how and why certain strategies are chosen, revealing the underlying principles that drive advanced operations in complex environments.

The narrative is built around three distinct operational tracks conducted within the same organization, which, when combined, enabled full compromise:

• CloudOps — decisions and paths that led to the compromise of cloud resources, including external dependencies and supply chain vectors;
• HybridOps — integration of physical, human, and logical layers, including physical intrusion and contextual exploitation of the corporate environment;
• FinancialOps — understanding of financial and operational processes that enabled the execution of validated fraud without generating meaningful alerts;

Across these three domains, the session demonstrates how cloud, physical, and financial layers can be exploited in a coordinated manner while maintaining consistency with expected system behavior.

As a central part of the session, a real operation conducted by our company will be presented, in which an organization was fully compromised through a coordinated, multi-layered approach. The scenario includes:

• exploitation of a 0day in a physical intrusion context, enabling continuous presence within the environment for 30 days without detection, including a 0-click account takeover scenario;
• progression from an initial phishing vector to full compromise of cloud assets, including resource abuse within well-mapped compromised accounts;
• analysis of critical business services and processes, resulting in confirmed financial fraud through the exploitation of a client-side trust model in a payment system, with direct reputational impact;

In this case, there was no single event that defined the attack, but rather a sequence of controlled and coherent actions that blended into normal operations, enabling full control without triggering traditional detection mechanisms.

The goal of this talk is to provide security professionals with a practical and structured understanding of how APTs actually operate, offering deeper insight into how decisions are made throughout a real-world operation.

The key takeaway is that APT operations are not effective because they are invisible, but because they do not behave like something that deserves attention.

People:
    SpeakerBio:  Jonathan Coradi

Jonathan Coradi is a RedOps Tech Lead at Hakai Offensive Security, with over 7 years of experience in offensive security. He has led offensive operations across industrial, financial, and banking sectors in Brazil, specializing in advanced penetration testing, Red Team simulations, and physical intrusion engagements. Jonathan is also an elite Bug Bounty Hunter — currently ranked Top 1 on BugHunt — and has reported critical vulnerabilities to companies like Microsoft, Uber, and Mercado Livre, among others.

SpeakerBio:  Oliveira Junior

Oliveira Lima is the founder of Hakai and has over 15 years dedicated to the field of cybersecurity, focusing on penetration testing and Red team operations. As a researcher, he has reported numerous vulnerabilities in large companies such as Trend Micro, CISCO, Dlink, etc. Additionally, he has registered more than 40 CVEs. Oliveira continues to be part of the team leading Red team operations.

Creator: Bug Bounty Village

At Google, we receive hundreds of reports per week (not exaggerating) that claim to identify vulnerabilities in our AI products. However, only a tiny fraction of those (<1%!) will receive a reward. But there is hope! When we break down the reasons so many reports are rejected, nearly all of them fall into a few common categories that make them invalid. Join the Technical Lead of Google’s AI Vulnerability Reward Program to walk through these common pitfalls, learn how to bug hunt more effectively on AI products, and hopefully, take home more bounties for your efforts. Outline I. Introduction Self intro, discuss the interest in AI VRP, introduce the problem of low success rates for researchers II. Overview of Google AI VRP – Brief program history, overview of triage/reward process – Explain VRP scope/rewards: In-Scope: Security vulnerabilities (Rogue Actions / Sensitive Data Exfiltration); Some AI Abuse categories Out-of-Scope: “AI slop,” simple alignment bypasses, or jailbreaks that do not result in a direct security impact on user data. III. Common Mistakes – Jailbreaks/Safety: Getting an LLM to generate “bad” content is not a security vulnerability we reward through the program (please report in-product!). – Self-pwns: Reports are over-reliant on social engineering or have unconvincing attack scenarios – Ignoring Context: Misunderstanding “sensitive data” versus expected model behavior. IV. Better Bug Hunting – Direct and Indirect Prompt Injection: Reviewing (un)successful AI VRP reports – Our programs look for indirect prompt injection attacks; this isn’t always easy to understand – Note: This section will include details from a recently rewarded and disclosed AI VRP report [details TBC pending researcher/corporate comms approval] – Clear and Actionable Reporting: – We want to help teams fix bugs quickly; you can help us do that by writing clear and actionable reports – Reproduction can be challenging due to non-determinism V. Conclusion / Q&A

People:
    SpeakerBio:  John Kotheimer, Senior Security Engineer, Google

John is a Senior Information Security Engineer and Technical Lead of the AI Vulnerability Reward Program (AI VRP) at Google in New York City. John has significant experience operating bug bounty programs–including a previous role as TL of the Abuse Vulnerability Reward Program at Google–as well as a background in infrastructure security, red teaming, and incident response. As AI VRP lead, John oversees the triage of hundreds of AI bug reports submitted to the program every week. John also coordinates the panel that determines rewards for accepted AI bugs at Google and helps plan and run many of Google’s bugSWAT live hacking events. Before joining Google in 2019, John was an information security consultant at Mandiant and completed graduate study at Carnegie Mellon University. Outside of work, he enjoys travel, craft beer, and riichi mahjong.

Creator: Middle Easterns & Africans in Cyber Security (MEACS)

As AI coding assistants accelerate software development, the volume of pull requests at Datadog has grown to nearly 10,000 per week, increasing the risk that malicious changes slip through due to review fatigue. To address this, Datadog built BewAIre, an LLM-powered code review system designed to identify malicious source code changes introduced by threat actors. By reducing approval fatigue for developers while increasing friction for attackers, BewAIre guides human reviewers to the areas where judgment matters most, without slowing developer velocity.

In this breakout session, Andrew Krug, Head of Security Advocacy and Research will share why BewAIre was built, how it evolved from a hackathon experiment into a production-grade internal system, and the key architectural decisions and trade-offs involved along the way. They will discuss what worked, what didn’t, and the limitations the team encountered when applying LLMs to security-critical workflows.

They will also cover how BewAIre is now being integrated into Datadog Code Security, and what it takes to turn an internal engineering tool into a product capability used at scale. Attendees will leave with practical lessons on building, hardening, and productizing LLM-powered systems and how you can use LLMs to minimize the security risks introduced by the new LLM-code-generation paradigm.

People:
    SpeakerBio:  Andrew Krug
No BIO available

Creator: Bug Bounty Village

Unicode exploitation does not end with normalization. Modern web applications process input through layered pipelines: URL decoding, UTF-8 validation, WAF transformations, framework parsing, surrogate handling, database collation, HTML entity decoding, and increasingly, LLM preprocessing. Each layer implements subtly different assumptions about character validity and equivalence. When those assumptions diverge, security boundaries fail. Building off our popular Black Hat USA 2025 Unicode Briefing, we have continued our research into the complex world of Unicode processing. This research exposes a new class of Unicode pipeline vulnerabilities that extend far beyond canonical normalization issues. We demonstrate how attackers can weaponize illegal UTF-8 sequences to break RE2 validation, exploit surrogate-to-replacement conversions (U+FFFD) to alter semantics, abuse hex overflows to generate filtered characters, and bypass Host/Secure cookie protections via Unicode whitespace desynchronization. We show how MySQL zero-weight collation rules enable filter bypasses even after normalization, how WAF/browser canonicalization mismatches lead to XSS and RCE (including analysis of CVE-2025-55182), and how invisible Unicode variation selectors can jailbreak LLMs or conceal supply chain malware. Using real-world telemetry, live demonstrations, tooling updates and hands-on lab environments, this Briefing reframes Unicode as a distributed parsing vulnerability class, not a character encoding footnote. Unicode is no longer just a normalization problem. It is an architectural attack surface.

People:
    SpeakerBio:  Ryan “ryancbarnett” Barnett, Senior Threat Research Manager, Akamai

Ryan Barnett is a Senior Threat Research Manager leading the Akamai App and API Protector (WAF) product. He also acts as s triager on Akamai’s and customers’ bug bounty programs. In addition to his primary work at Akamai, he is also a former Faculty Member for the SANS Institute, a WASC Board Member and OWASP Project Leader for: ModSecurity Core Rule Set (CRS) Web Hacking Incident Database (WHID). Mr. Barnett has also authored two web security books: Preventing Web Attacks with Apache (Pearson) and The Web Application Defender’s Cookbook: Battling Hackers and Defending Users (Wiley).

SpeakerBio:  Isabella “4ng3lhacker” Barnett

Isabella Barnett is a Software Engineering Intern at Akamai and a junior at George Mason Honor’s College studying Cyber Security Engineering.

Creator: Red Team Village

For years, Red Teams have relied on techniques like NPPSpy and LSASS dumping for credential harvesting. But as EDRs and defenders get smarter, these traditional paths are heavily monitored and immediately flagged. What if we could intercept credentials right at the source—the Windows Logon UI—while seamlessly bypassing Application Control solutions?

In this talk, we will journey deep into the arcane world of the Windows Credential Provider Framework and COM (Component Object Model) architecture. I will demonstrate how to build a stealthy “Middle-Man” Credential Provider that wraps around the default Windows Password Provider. By abusing interfaces like ICredentialProviderFilter and intercepting methods like GetSerialization, we can silently harvest both cleartext passwords and serialized authentication payloads.

We’ll explore the COM “chicken-and-egg” instantiation process, the operational security considerations of operating inside logonui.exe, and most importantly, how this technique achieves a complete bypass of AppLocker by proxying execution through a trusted system process. Finally, we’ll discuss its limitations against strict WDAC policies and how defenders can hunt for this elusive persistence mechanism.

Research Motivation & Core Concept

The core insight of this research is moving away from the highly monitored Network Provider registry keys (HKLM\SYSTEM\CurrentControlSet\Services\LanmanWorkstation\NetworkProvider) and LSASS memory space. Instead, this research leverages the Windows Credential Provider Framework. By writing a custom COM DLL, we can create a “wrapper” around the native Windows Password Credential Provider ({60b78e88-ead8-445c-9cfd-0b87f74ea6cd}).

To avoid drawing suspicion, the user must not see two password fields. This research utilizes the ICredentialProviderFilter interface to explicitly filter out (hide) the default Windows provider, leaving only our malicious wrapper.

Technical Insights & Evasion Mechanics

1. COM Interface Abuse: The talk details the exact implementation of IUnknown, IClassFactory, ICredentialProvider, and ICredentialProviderCredential required to successfully trick logonui.exe into loading the malicious provider.
2. Data Harvesting: I will break down how to extract cleartext passwords by intercepting the SetStringValue method, and how to capture serialized payloads (like NTLM hashes/Kerberos material) by intercepting GetSerialization before passing execution back to the native provider.
3. AppLocker vs. WDAC: A major finding of this research is Application Control bypass. Because the custom DLL is loaded by logonui.exe (a highly trusted, system-level process), AppLocker is completely bypassed. However, I will also transparently discuss WDAC: under strict Enforce Mode, WDAC’s CodeIntegrity checks will successfully block the unsigned DLL (Event 3033), making this a great comparative study of AppLocker vs. WDAC for defenders.
4. OpSec Constraints: I will cover the developer-side OpSec required, such as compiling with /MT (statically linking the CRT) to avoid MSVCP140.dll dependency crashes in the logon screen, and securely saving harvested credentials via XOR encryption to C:\Windows\Temp.

Target Audience

• Red Teamers looking for stealthy persistence and credential harvesting techniques;
• Blue Teamers and Security Engineers tasked with tuning EDRs and writing Application Control (AppLocker/WDAC) policies.

People:
    SpeakerBio:  Sohail Saha

Sohail is a Penetration Tester at Optiv specializing in Windows internals and offsec tool development. Previously a developer at Polygon, they now leverage their software engineering background to build offensive tooling and research OS architecture. When not bypassing security controls or writing C++, he can be found playing guitar, reading, or working on their debut sci-fi novel.

Creator: Red Team Village
Abstract:

Capture The Flag competitions are often dismissed as “just games,” but modern Product Security teams increasingly rely on the exact skills they develop: adversarial thinking, vulnerability analysis, exploit chaining, and secure design under pressure. Whether you are an aspiring security professional or an experienced offensive security practitioner, in this session, you will learn a structured CTF-driven learning roadmap for developing practical security skills and will learn how hands-on, real-world customer-facing offensive engagements contribute to building deep product security expertise. We will also cover how AI capabilities are accelerating this journey of learning and securing the systems. Attendees will participate in a hands-on exercise to explore the challenge through the lenses of CTF problem solving, penetration testing, and Product Security.

===================================================

Introduction: (1 minute)

Session Overview: (2 mins) • During this session, you will see a practical framework to become a trusted partner in building secure applications rather than being a gatekeeper. Starting with a practical framework of learning and approaching CTFs, to using it as an offensive security practitioner to develop stronger application security capabilities. We also see how AI can accelerate and reshape modern security workflows. • In this session, attendees will be given a challenge designed to get hands-on experience of solving the CTF, creating an offensive security report, and sharing recommendations from a product security engineer’s perspective to secure the applications/systems. After the session, I will review each participant’s approach and award swag to the top performers.

Intended Audience: (2 mins) • Aspiring security professionals • Experienced penetration testers • Security engineers • Students entering cybersecurity

Capture the Flag (7 minutes) • What is CTF? • Why is CTF crucial? • Hands-on experience • Risk-free environment • Collaboration and teamwork • Skills that CTFs actually teach • Avoiding rabbit hole • Structured approach to learn concepts • OWASP Top 10 • Domains: APIs, Web Application, Crypto, Cloud • A few common vulnerability areas to consider: Injection, File Upload Vulnerabilites, AuthN, AuthZ, Remote Code Execution

Penetration Testing (7 minutes) • What is penetration testing? • Map skills learned in CTF to organizational context using different methods (black box, white box, and grey box testing) • What is the focus area: • Effective technical reports that include: •Details on the process: environment considerations, recon, scanning, and exploitation • List of Findings with details about: What exactly is the specific finding, Severity, Impact, How to remediate the finding

• CTF becomes pentesting when you stop asking: • Can I exploit this? • Where can I find the flag? • And start thinking: • What does this mean for customers? • How does it impact their product?

• Note: One starts to build customer trust by helping them with the business context of findings and how to reduce the risk

Product Security Engineer (7 minutes) • Shift in mindset from “find the bug” to “how do we prevent this everywhere? With the knowledge gained from CTF and real-world pentesting.” • Holistic approach of shifting left the security: • Threat Modeling • Implementing Security Best Practices: • Integration of the SAST tool into the CI/CD pipeline • Logging and monitoring • Cloud Configuration reviews • Least privileges • Vulnerability Management Comparison • Compare with the Software Engineer journey to Programming -> Data Structures -> Front End Developer -> Full Stack Developer

Resource: (2 minutes) • Ask the audience for any recommendations that helped them be strong at what they do • Provide resources that I found useful

Recap: (2 minutes) [CTF – “How do I break this to get the flag?” → Pentest – “How bad is this?” → Product Security Engineer – “How do we prevent this everywhere?”] → AI – “How do we move faster without losing depth?”

Task (2 minutes): • Give them details about the challenge environment • Explain to them the expectations • Find the flag • Penetration Test Report • Explain their approach to secure the feature (Secure Design review, SAST recommendation, how would you prevent it across the application?) • The winner will get swag

Key Takeaways: (3 minutes) • Security is not just: • Finding security gaps • Shipping reports • Security engineers become trusted partners when they: • Know the attacker’s mindset • Have hands-on experience in ethical hacking • Understand product goals • Understand how systems fail • Gives practical guidance by being part of secure development with developers • Communicate risk clearly • Improve application security at scale

Q&A (5 minutes)

People:
    SpeakerBio:  Drew Thompson

Drew Thompson is a Principal Consultant at UltraViolet Cyber, where he specializes in cybersecurity training, offensive security, and AI-enabled security practices. He designs and delivers hands-on training for security practitioners, develops technical enablement programs, and works closely with consultants and customers to translate emerging attack techniques into practical defensive capabilities. Drew is passionate about making complex security topics accessible through real-world demonstrations, interactive labs, and practical research.

SpeakerBio:  Monish Alur Gowdru

Monish Alur Gowdru is a cybersecurity professional with over 6 years of experience in application security and software development. He enjoys helping product teams to secure applications end-to-end throughout the software development lifecycle. He actively contributes to the cybersecurity community as a speaker, mentor, and judge. His ongoing involvement includes prominent work with DEF CON and BSides Edmonton.

LinkedIn: https://www.linkedin.com/in/ag-monish/

Creator: Adversary Village

Red Teaming is about thinking and acting like an attacker to improve an organization’s defenses. In practice, it is less about flashy exploits and more about providing prioritization signals, validating and improving detection capabilities, and telling compelling stories that drive meaningful security outcomes.

The rise of AI has introduced a new challenge and a new mandate. Organizations are increasingly asking Red Teams to “use AI to hack things” as both attackers and defenders race to understand what AI can and cannot do. To cut through the hype, the role of Red Teams to speak truth to power is more important than ever.

Our responsibility is to separate speculation from reality. To understand the capabilities of AI-powered attackers, assess how well our defenses stand up against them, and evaluate the risks introduced by new AI-driven attack surfaces. By doing so, we help organizations make informed decisions about where to invest, what to defend, and how to prepare for the threats that matter most.

People:
    SpeakerBio:  Michael Leibowitz, Senior Principal Troublemaker, Oracle Red Team

Michael (@r00tkillah) has done hard-time in real-time. An old-school computer engineer by education, he spends his days hacking the mothership for a fortune 100 company. Previously, he developed and tested embedded hardware and software, fooled around with strap-on boot roms, mobile apps, office suites, and written some secure software. On nights and weekends he hacks on electronics, writes CFPs, and contributes to the NSA Playset.

SpeakerBio:  Niranjanaa Ragupathy, Security Engineer Manager, Google Red Team

Niru is a Security Engineer and Manager at Google. She leads the Offensive Security team, which focuses on hacking Google to improve its overall security. In her free time, Niru enjoys doodling and creating Capture The Flag (CTF) challenges.

Creator: Bug Bounty Village

Web cache poisoning is often written off as a class of vulnerabilities with limited real-world impact. Most reported instances only expose non-sensitive user metadata, cause temporary, minor disruptions to static content, or require such narrow, impractical conditions to exploit that they rarely move beyond theoretical proof-of-concept. For TikTok’s Vulnerability Management team, this framing shaped our initial approach to triaging cache poisoning submissions for years—until two radical researcher submissions upended that assumption entirely, proving misconfigured web caches can undermine every pillar of the CIA triad (confidentiality, integrity, availability) to enable catastrophic, scalable harm. In this talk, we’ll start with a foundational breakdown of web cache poisoning: common misconfigurations (from flawed cache key logic to mishandled origin headers) that enable exploitation, and the limited impact profiles that led our team (and many in the bug bounty ecosystem) to historically deprioritize these flaws. We’ll then deep dive into the game-changing submissions that reshaped our security posture and why security teams and bug bounty hunters must re-evaluate how they assess cache poisoning risk—moving past surface-level assumptions about limited impact to audit for hidden, critical exploit pathways. Whether you triage web vulnerabilities, build global web infrastructure, or hunt for bugs at scale, this session will equip you to spot and remediate cache poisoning risks before they’re weaponized against your users.

People:
    SpeakerBio:  Glendon Chong, Tiktok

I am part of the Vulnerability Management team for TikTok. Over the past year, I’ve been actively involved in web application vulnerability triage, collaborating with security researchers and internal teams to dissect, validate, and remediate a wide spectrum of web-based threats. My work centers on bridging reporter expertise and TikTok’s security posture—including in-depth negotiations over CVSS scoring, refining triage workflows for emerging attack vectors, and translating complex vulnerability details into actionable remediations. I bring hands-on experience identifying gaps in modern web architectures, advocating for fair, transparent risk assessment with the bug bounty community, and aligning vulnerability prioritization with our platform’s commitment to user safety.

Creator: IoT Village

Kindles, Kobos, Boox and BigMes there’s no shortage of eReaders to choose from in 2026, with their paper-like eInk displays designed for one thing: reading books. But under the surface of these minimalist devices is a surprisingly hackable device. From Kindle jailbreaks, to Android apps, to flashing custom firmware. We’ll take that dust-gathering device off of your nightstand and onto your lab bench to talk about the vulnerabilities and customisability of these devices.

People:
    SpeakerBio:  Katie Paxton-Fear, Security Advocate, Semgrep

i used to make applications and now I break them, hacker, bug bounty hunter, and educational YouTuber.

Creator: Blacks In Cyber Village

This thought-provoking session examines how bias in AI systems used across cybersecurity for fraud detection, identity systems, and behavioral analytics can inadvertently create structural weaknesses that adversaries might exploit. We will explore how bias can be introduced through data collection, labeling, and modeling, leading to uneven false positive and false negative rates across different populations. Understanding these inconsistencies is crucial, as they can offer attackers avenues to map detection thresholds and bypass security systems. The session will also connect these security risks to emerging regulatory frameworks like GDPR and the AI Act, demonstrating how compliance needs are aligning with core security practices. Join us to learn why integrating equity into threat modeling, red teaming, and system auditing is essential for robust cybersecurity, moving beyond treating fairness as an afterthought.

People:
    SpeakerBio:  Dr. Fatou Sankare, Cybersecurity Professional / Community Speaker

Dr. Fatou is a Cyber Engineer and an adjunct professor, dedicated to increasing cyber education, particularly in underestimated communities, through Datacation LLC, which they founded. They are a Certified Ethical Hacker and hold the AWS Machine Learning Specialty Certification.

Creator: Cryptocurrency Village

In this hour, we bring together international hackers to discuss cutting-edge techniques in cryptocurrency security. Our experts break down how distributed red and blue teams face off against crypto developers to secure stablecoin vaults — and how they translated that daily work into a format suited for the Cryptocurrency Village at DEF CON.

We’ll also explore how classic Opensource intelligence (OSINT) techniques are converging with modern financial technology to unmask bad actors and bypass security mechanisms. Finally, the panel reveals the tricks that black hats use to extract unauthorized value from real-world mainnet blockchains, including Bitcoin, Solana, Ethereum, and Monero.

As a bonus, you’ll get a first look at our forthcoming DEF CON and Black Hat training course, Breaking and Defending Cryptocurrency. We’ll share our design process, plans, and the challenges of building a half-dozen hands-on activities that teach the ins and outs of offensive security as it applies to blockchain-based financial technology.

People:
    SpeakerBio:  Dinmukhammed “Celestial” Kabiden, OSINT Mindset

Dinmukhammed Kabiden, also known as Celestial, is a CTI analyst and Web3 investigator from Kazakhstan focused on OSINT, blockchain investigations, crypto fraud, and cyber threat intelligence. He has experience investigating fake airdrops, wallet-drainer activity, on-chain fund flows, scam infrastructure, and OPSEC mistakes that expose threat actors. He is also the organizer of OSINT Mindset and speaks about practical Web3 investigations, intelligence reporting, and the intersection of off-chain and on-chain evidence.

SpeakerBio:  Izdihar, Swarmnetics / cyber673 / Brunei Cyber Security Association

Izdihar is a cybersecurity practitioner who works in penetration testing and tinkers with a homelab in his spare time, where he has been experimenting with running blockchain testnet nodes. His interest in cryptocurrency is mostly about understanding how the infrastructure works. He helps run cyber673, a small grassroots community in Brunei.

SpeakerBio:  KL4R10N

Sarthak Taneja is a security professional specializing in detection engineering, threat intelligence, cloud forensics, and incident response across fintech, crypto, and Web3 environments. His work focuses on securing digital asset infrastructure, custody workflows, cloud-native systems, and blockchain-adjacent attack surfaces, with experience investigating advanced threat activity from groups such as Lazarus, Kimsuky, BlueNoroff, and ShinyHunters. He has led initiatives across SOC maturity, Detection-as-Code, zero-trust controls, and purple teaming, and has spoken at international security conferences including DEF CON.

Creator: Red Team Village

BloodHound is great… until your VM dies under Neo4j load, your login expires mid-exam, or you need hybrid AD + Entra ID insights without spinning up servers. Enter BloodBash: a standalone Python tool that ingests SharpHound (v6+) and AzureHound JSON directly, builds a full NetworkX graph in-memory, and spits out attack paths, misconfigs, and abuse suggestions — all offline, in colorful terminal glory. Born from a real OSCP crisis (BloodHound failed hours before exam start → BloodBash saved the DA path), this “vibe-coded” tool now detects shortest paths to high-value targets, ADCS ESC1-8 vulns, RBCD, Kerberoasting, GPO abuses, exposed app secrets/certificates, MFA bypass risks, privileged Entra roles, and unified hybrid attack chains. In this talk, you’ll see: The OSCP war story that sparked it Live demo: Drop SharpHound + AzureHound files → instant colorful output, path visualization, abuse commands Key detections walkthrough (on-prem + cloud) Why lightweight CLI > GUI bloat for field ops, labs, and quick checks Roadmap: Metasploit integration, more exports, community features Full open-source on GitHub (https://github.com/DotNetRussell/BloodBash) — star it, fork it, break it. Attendees walk away with a new go-to tool for AD/hybrid recon.

People:
    SpeakerBio:  Anthony Russell, Cyber Security Software Engineer

Anthony Russell, is a senior cyber security engineer with over 13 years of professional experience building software. He has a focus in information security and has been featured in 2600 magazine, on Hak5 and has spoken at both Defcon and DerbyCon multiple times. Favorite things to discuss are blockchain technology, baking custom IoT devices, and everything infosec. You can see more of Anthony’s work at SquidHacker.com or Twitter.com/DotNetRussell

Creator: Red Team Village

In this session, attendees will learn how to design data models and build extensions for BloodHound OpenGraph.

After a presentation where I will go over the core concepts & components of BloodHound OpenGraph and the thought process behind BloodHound data modeling, attendees will design a model and build an extension from scratch based on a provided dataset.

I you like to tinker with BloodHound and are curious about the new OpenGraph features, this session is for you…

Note: Bring your own laptop with BloodHound Community Edition installed

People:
    SpeakerBio:  JD D

WHOAMI

Creator: Red Team Village

In this session, we’ll give a short demo of the BloodHound OpenGraph platform, and show off some modeling capabilities and how to get started working with it. We will show participants an example using lab data with a fake AD environment and Nessus/Vulnerability data and demonstrate how to integrate other sources of data into the existing graph and how easy it is to get started.

People:
    SpeakerBio:  Rohan Vazarkar

Former pentester and original author of BloodHound, turned developer.

SpeakerBio:  Wes Miller

Wes Miller is on the Developer Relations team at SpecterOps, and has worked on many of the core components of BloodHound Community Edition including many of the core features of OpenGraph

Creator: Red Team Village

BloodHound Based CTF

People:
    SpeakerBio:  Hugo van den Toorn

Hugo is former Chief Information Security Officer and has now transitioned back to help other organizations understand adversary tradecraft. With over twelve years of experience in the Information Security industry, he has a solid technical and executive background as hands-on security leader.

Hugo has experience with and a keen interest in Social engineering, phishing and physical penetration testing. Nowadays, Hugo takes pride and joy in helping individual team members and the business grow. With a strong technical foundation, Hugo combines his passion for security, teaching and hacking with a drive for continuous improvement and optimization of people, processes and technology.

Creator: Blue Team Village

Join BTV staff as we share this year’s highlights, puzzle and Obsidian Capture the Flag event stats and winners, and shout our amazing volunteers, staff, sponsors, and of course attendees. Say goodbye for now, and snag any leftover swag!

Creator: Payment Village

Buy Now Pay Later has grown from a niche convenience to a $560 billion global ecosystem used by approximately 380 million people. Klarna, Afterpay, Affirm, and PayPal Pay Later process billions in transactions annually — and every transaction flows through a series of REST API calls. Security research on BNPL has focused entirely on fraud: account takeover, synthetic identity, first-party fraud. No one has looked at the API layer. This talk does.

We systematically test the BNPL payment flow — session creation, authorization, order management, and capture — using public sandbox environments. Our methodology focuses on business logic flaws: scenarios where the API functions exactly as designed but can be abused in ways developers never anticipated.

Confirmed finding: Klarna’s capture endpoint accepts any captured_amount value down to 1 cent ($0.01) on a fully authorized order, with no minimum validation. A $100 authorized order can be captured for $0.01 — returning HTTP 201 Created with Klarna-Order-Validation: Valid. In vulnerable merchant implementations, customers receive goods while only $0.01 is collected — a $99.99 financial loss per transaction.

Attendees will leave with a practical attack methodology for BNPL integrations, an open-source reference implementation (vulnerable + patched), and a merchant-side security checklist.

People:
    SpeakerBio:  Furkan Fatih Demir, Independent Security Researcher & Penetration Tester, Barikat Cybersecurity

Conducts authorized security assessments across financial services, aviation, public sector and education, focused on business-logic vulnerabilities that evade conventional tooling.

Creator: Game Hacking Village

People:
    SpeakerBio:  InsiderPHD
No BIO available

Creator: La Villa Community

Los bootkits UEFI continúan siendo una de las capacidades más avanzadas utilizadas por grupos APT modernos. Su capacidad para ejecutar código antes del sistema operativo, obtener persistencia a largo plazo y operar por debajo de muchas soluciones de seguridad los convierte en una herramienta extremadamente valiosa tanto para atacantes como para equipos de Red Team que buscan emular amenazas reales.

En esta sesión mostraremos cómo utilizar un bootkit UEFI para reproducir cadenas completas de ataque observadas en operaciones reales. Analizaremos el recorrido desde el firmware hasta el kernel de Windows, incluyendo persistencia y despliegue de componentes adicionales de ransomware en el kernel.

La charla incluirá una demostración en directo en la que un sistema Windows moderno será comprometido mediante un bootkit UEFI, mostrando paso a paso cómo este tipo de implantes se instalan y permiten reproducir capacidades observadas en operaciones reales de grupos APT. Además, se compartirán los recursos y proyectos utilizados durante la investigación para que otros equipos de seguridad puedan reproducir los escenarios presentados en sus operaciones diarias.

People:
    SpeakerBio:  Alejandro Vázquez Vázquez, Independent Security Researcher

Alejandro Vázquez Vázquez es un especialista en técnicas de seguridad ofensiva de bajo nivel, con foco en Red Teaming, Windows Internals y UEFI. A lo largo de su trayectoria ha participado en operaciones de Red Team y emulación de adversarios, además de desarrollar capacidades ofensivas orientadas al estudio y reproducción de técnicas utilizadas por grupos APT modernos. Su trabajo diario se orienta a la creación, análisis y despliegue de implantes capaces de operar por debajo de los mecanismos de seguridad modernos, abarcando desde técnicas pre-boot hasta módulos en kernel y componentes de usuario para persistencia.

Comparte su conocimiento siendo docente en diversos másteres de postgrado de universidades y entidades privadas, donde imparte formaciones en ingeniería inversa, análisis de malware, desarrollo de exploits y desarrollo de herramientas para evasión de soluciones antimalware, acercando la seguridad ofensiva a las nuevas generaciones de profesionales.

Colabora con diversas comunidades centradas en áreas de low-level, como Off By One Security y grupos de DEF CON, conferencia en la que ha sido ponente principal presentando los proyectos Abyss (UEFI Bootkit) y Benthic (Windows Kernel Rootkit). A lo largo de su desarrollo profesional también ha participado en iniciativas de investigación y desarrollo relacionadas con firmware, metahoneypots, análisis de amenazas y plataformas destinadas al estudio de actores maliciosos.

Creator: Bug Bounty Village

AI has landed on both sides of the bug bounty table at once. Hackers are being told to use it, programs are drowning in the slop it produces, and a whole new attack surface, agents, RAG pipelines, and tool-calling LLMs, just showed up in scope.

This panel brings together working hackers who run AI in their daily flow, LLM security researchers who break AI for a living, and a moderator who has spent a decade in the scene. In 45 minutes they cut through the hype and get specific: which parts of recon, review, and reporting AI actually handles, how to hack the LLM-backed apps now in scope and get paid for it, and why the real edge isn’t ChatGPT but the private agents and skills hackers are quietly building to turn one good idea into a hundred bugs.

People:
    SpeakerBio:  Ben “NahamSec” Sadeghipour

Ben Sadeghipour (NahamSec) is a security researcher, ethical hacker, and educator who has spent more than a decade finding and disclosing critical vulnerabilities in some of the world’s largest organizations. As one of the most recognized names in the bug bounty community, he has reported thousands of security flaws and consistently ranked among the top researchers on leading hacking platforms. Beyond his own research, Ben is passionate about lowering the barrier to entry in cybersecurity. Through his widely followed educational content, live streams, and the conferences he founds and organizes, he has helped train a new generation of hackers around the world. His work blends deep technical expertise with a commitment to mentorship and community building.Ben speaks regularly at industry conferences on offensive security, bug bounty hunting, and building a career in cybersecurity.

SpeakerBio:  Vitor “busf4ctor” Falcao Habibe Costa, Frontier AI Red Team Operator, BT6

Vitor is an AI security researcher, Community Manager at Critical Thinking, and a Google VRP hunter, named Best AI Researcher at Google bugSWAT. He’s part of the BT6 team and hunts bugs full time.

SpeakerBio:  Joey Melo

Joey is a Principal Security Researcher at CrowdStrike and a contributor to AI safety programs at OpenAI and Anthropic. He specializes in offensive security research and adversarial analysis of complex systems, with a particular focus on AI/ML infrastructure and large language models. His work has led to the discovery of critical vulnerabilities and novel exploitation techniques, helping organizations better understand and defend against sophisticated threat actors. He has experience across red teaming, exploit development, and designing resilient architectures for high-risk environments. Joey has earned recognition through bug bounty programs and multiple security competitions, with over $100,000 awarded for vulnerability research. He is also an OSCP, OSCE³ and CRTO-certified professional.

SpeakerBio:  Dustin “ph1r3574r73r” Farley, BT6

Dustin Farley has spent more than a decade building software, securing it, and figuring out how to break it. Today, he focuses on AI security, where he spends his time red teaming large language models, AI agents, and other emerging AI systems. His interests include prompt injection, jailbreaks, adversarial testing, and turning weird edge cases into useful security research. He firmly believes that the fastest way to understand a system is to see how it fails.

SpeakerBio:  Ads Dawson, Staff AI Security Researcher, OWASP GenAI Security Project founder

Ads Dawson founded the OWASP GenAI Security Project and led it to flagship status — the fastest in OWASP history. As a Staff AI Security Researcher at Dreadnode, he specializes in exploiting ML and AI systems, harnessing AI for offensive cybersecurity through capability development and exploit development, and conducting red team operations against frontier models for government, military, and tier-1 labs. He is lead author of AIRTBench (arXiv), the first benchmark for autonomous AI red teaming in LLMs, and author of AI Native LLM Security (Packt, 2025).

A senior red team operator with BT6 (the frontier AI red team), ranked #2 in Canada on HackerOne (2025/2026), and selected for Meta’s MBBRC live hacking event, Ads is a HackerOne US South Ambassador, BugCrowd Hacker Advisory Board member, MITRE AI Working Group contributor, and leads the OWASP Toronto chapter. He spoke at Bug Bounty Village DC33 on the BT6 AI jailbreaking panel and is also presenting “Exfil Everything: A Year of Stealing Data from AI Agents” at Bug Bounty Village DC34 (schedule pending publication).

SpeakerBio:  Johann “wunderwuzzi23” Rehberger
No BIO available

Creator: Red Team Village

NEURO is a realistic enterprise AI finance assistant built with an LLM gateway, RAG/vector search, tool calling, MCP integrations, guardrails, audit logs, and a downstream exfiltration simulation.

In this hands-on tactic, attendees will attack and defend the same AI stack from both red-team and blue-team perspectives in our custom CTF.

Participants will work through practical compromise paths including prompt injection, retrieval poisoning, unsafe tool use, MCP trust boundary abuse, and data exfiltration patterns. They will then inspect evidence, review guardrail and audit signals, and apply defensive controls using Cisco Secure AI capabilities such as inspection,scanning, and runtime policy hardening.

The goal is to show how modern AI systems can fail as complete stacks, not just as chat prompts, and how defenders can reason about the full chain from user input to retrieval, tools, model behavior, telemetry, and response controls.

The tactic is designed to fit a 30-120 minute wave with small groups and can be repeated for multiple attendee groups.

People:
    SpeakerBio:  Jason Ludwig

Jason is a Principle Solutions Architect with over 20 years of experience in software engineering and AI development. He specializes in advanced machine learning systems over many domains, including smart-camera AI for industrial safety, real-time predictive analytics, sentiment analysis, LLMs/Rag, and AI cybersecurity platforms. His work has been featured in Wired, Time, and TED. Jason has successfully led teams at top global organizations like World Wide Technology, Mastercard, and Monsanto, driving the delivery of large-scale, AI-enabled solutions.

Creator: Payment Village

The analysis begins at Ring -3, advancing toward the positive rings. Starting from the on-screen user

perspective imposes too many limitations. The shift in approach was deliberate: rather than assuming a

conventional attack vector such as connecting an external device or performing a kiosk-level bypass the

decision was made to start from the deepest layers of the system. This compromise is detected for the

following reason:

• DMA protection exists at the BIOS level.

The goal of this compromise is to activate PCI/PCIe through BIOS in order to perform a DMA

This section will be referred to as “BIOS Unprotection.”

How was this achieved?

Two methods were identified:

A bridge was found that triggers a BIOS configuration reset.

BIOS duplication with encryption validation bypass (enables both Downgrade and Upgrade).

DMA or complete chain will not be shown since it is a delicate topic but BIOS research will be deepened

since it does not only apply to ATMs.

People:
    SpeakerBio:  Arnold Jared Morales Yepez, Senior Team Lead, Grupo Salinas

Self-taught in computer security since age 12; holds a degree in Computer Forensics and Cybersecurity and is pursuing a Master’s in AI and Cybersecurity.

—

Desde los 12 años, me he dedicado a la informática con un enfoque especial en la seguridad. Actualmente, a mis 22 años, sigo explorando este mundo con la misma pasión, impulsado por la constante evolución de la tecnología y su interminable curva de aprendizaje.

Cuento con una carrera en Cómputo Forense y Ciberseguridad, actualmente curso una maestría en Inteligencia Artificial y Ciberseguridad.

Certificaciones: CWEE | CAPE |CPTS | EWPTX | CRTO | eMAPT | OSCP | OSCP+ |CEH V13 | EJPT| HTB prolabs Hades – Cybernetics – Zephyr – APTlabs | MDK

Creator: Cryptocurrency Village

Zero-knowledge proofs are a method of demonstrating that a statement is true, without revealing any other information about it, and form the backbone of all the cryptocurrencies we know and love. In this workshop, we will unpack what zero-knowledge proofs really are, how they work, and what they do to ensure that your personal information remains private. A familiarity with basic cryptography is helpful, but not strictly necessary.

People:
    SpeakerBio:  Luke Szramowski, Stack Wallet

Luke Szramowski is a mathematical researcher, with a Bachelor’s Degree in Mathematics and two Master’s Degrees, one in Math, with a focus in Number Theory and another in Math with a focus in Coding Theory. In his free time, Luke works on a litany of different math problems, mainly regarding Number Theoretic conjectures and playing all different types of games.

SpeakerBio:  Freeman Slaughter, Cypher Stack

Freeman is a cryptographic researcher specializing in zero-knowledge proofs and code-based cryptography. He has co-authored several influential peer-reviewed works, including designing the post-quantum signature scheme CROSS, which is currently under review for government standardization. His work is driven by a strong commitment to advancing the privacy and security of cryptocurrencies such as Monero and Salvium.

SpeakerBio:  Diego “rehrar” Salazar, Cypher Stack

Diego ‘rehrar’ Salazar has been around the FOSS and cryptocurrency communities for eight years. He owns and runs Cypher Stack, a company that performs novel research and makes contributions to various FOSS projects. He has organized and managed several villages at defcon, c3, and more.

Creator: Physical Security Village

You’ve cloned the keycards, you’ve bypassed the locks, and you’ve gotten into the building. Now what?

Convincing people to actually fix the security issues you find is always an uphill battle. It’s even worse in the world of physical security, where installing new locks costs more than deploying a software patch. This talk will explore how to make your physical red teaming activity scare the $hit out of your executives. We’ll establish persistence, pivot to the network, steal everything that isn’t nailed down, and put dollar amounts to the impacts of your findings.

Along the way, we’ll look at some of the techniques used by real-world threat actors that use physical attacks as part of their attack chains, and how you can effectively emulate the most realistic threats to your organization. Your adversaries don’t stop after getting in, and neither should you.

People:
    SpeakerBio:  Andrew Lebedinsky, Security Engineer, Red Team at Microsoft

Andrew Lebedinsky is a security engineer on the Microsoft Red Team. Their fields of interest include covert entry, radio hacking, and digital countersurveillance. In their spare time they can usually be found playing Source Engine-based FPS games and overthinking what to put in “speaker bio” fields.

Creator: Red Team Village

Modern MCP deployments assume that low-privilege tools remain low-risk when reviewed independently. In practice, host agents compose outputs across multiple tools and privilege levels, creating authority paths that do not exist in any single server’s standalone review.

This workshop demonstrates a compositional offensive primitive in MCP environments: a low-privilege server influencing the parameters of a high-privilege destructive operation through shared planning context.

Attendees will see a controlled lab reproduction where a read-only MCP server shapes the parameters of a privileged SQLite-backed delete operation. The approval prompt shown to the operator accurately describes the final action but omits the upstream provenance that influenced it.

The workshop introduces Urd, an open-source compositional analysis tool for MCP deployments. Urd reconstructs authority flow across servers using runtime tracing and manifest analysis, allowing operators and red teamers to identify undeclared cross-tool influence paths.

The focus of the session is not prompt injection or model jailbreaks. The focus is cross-server authority flow, provenance loss, and workflow-level trust failure in agentic systems.

This session includes a hands-on tactic where attendees reproduce the compositional failure locally, inspect runtime traces, generate divergence findings, and modify the lab to observe how low-trust inputs can influence privileged operations across MCP toolchains.

People:
    SpeakerBio:  Yevhen Pervushyn

Yevhen “valh4x” Pervushyn is the founder of Red Asgard, a security research firm focused on AI integration security, offensive security, and protocol analysis.

His work focuses on the operational security of agentic systems: orchestration boundaries, authorization flow, provenance, and workflow-level trust failure in AI deployments.

Prior to Red Asgard, he worked in blockchain security auditing and decentralized protocol analysis, experience that directly informs his current research into compositional trust failures in MCP-based systems.

He previously presented MCP security research at the SANS AI Cybersecurity Summit 2026. This session presents new technical research, tooling, and hands-on lab material focused on offensive analysis of MCP orchestration behavior.

Creator: Payment Village

Using decades-old OpenJDK PBKDF flaws, forgotten RFCs, hash collisions, and a few semantic edge cases, we’ll compromise multiple cryptographic systems with little more than a handful of carefully chosen values and several “useless” tools.

This talk explores how implementation bugs, legacy compatibility, and architectural assumptions can undermine otherwise sound cryptography—and why the real attack surface is often everything around the algorithm.

People:
    SpeakerBio:  Ken Pyle, Security Researcher, Bank of America

Cybersecurity researcher, exploit developer and conference speaker focused on vulnerability discovery, reverse engineering and adversarial cryptanalysis; has presented at DEF CON, ShmooCon and RSA Conference.

Creator: Nix Vegas Community

Privacy and digital agency are under siege by dark patterns, aggressive data harvesting, and SaaS monopolies. It’s time to reclaim our digital self-determination. This talk demonstrates how to build a production-ready, fully autonomous personal cloud using the declarative power of NixOS.We will explore how to self-host drop-in replacements for mainstream services—including Nextcloud for storage, Vaultwarden for passwords, and Matrix for communication. Attendees will learn how to turn any old PC or cheap VPS into an unbreachable digital fortress featuring automated backups, end-to-end encryption, and complete data isolation. Discover how NixOS makes personal infrastructure reproducible, resilient, and entirely yours.

People:
    SpeakerBio:  Maksim Kuskov

Security engineer in Yandex, CTF organizer & player @ HSE IS ’29

SpeakerBio:  Mikhail Ilin

I am an appsec with a strong background in competitive hacking. Over the years, I have won multiple national Cyber Security Sports Championships and currently serve as an ambassador for the Standoff 365 cybersecurity ecosystem. Beyond traditional technical exploitation, my research focuses on the intersection of clinical psychology and practical social engineering. As a strong advocate for digital self-determination, I view open-source infrastructure and NixOS not just as engineering tools, but as essential defensive shields against platform lock-in and corporate surveillance.

Creator: Cloud Village

Oracle Cloud Infrastructure (OCI) is arguably one of the lesser-explored major cloud platforms from an offensive security perspective. While OCI shares many familiar IAM concepts with AWS and GCP, its identity architecture—including sentence-based IAM policies, identity domains, dynamic groups, compartment hierarchies, and cross-tenancy permissions—creates authorization relationships and privilege escalation paths that benefit from OCI-specific analysis.

To help tackle these challenges, I built OCInferno, an open-source OCI reconnaissance framework for offensive security assessments. Alongside oci-lexer-parser, an ANTLR-based parser for OCI’s sentence-based IAM policies, and OCISigner, a Burp Suite extension for transparently signing OCI API requests, a pentester/blue teamer can automate enumeration, model OCI IAM relationships, and build OpenGraph-based attack paths similar to BloodHound to help visualize privilege escalation opportunities.

This talk will demonstrate how to model OCI IAM relationships and uncover OCI-specific attack paths that are difficult to spot from policy text alone. It will also cover a security issue I uncovered that was acknowledged by Oracle in the June 2026 Critical Security Patch Update under its Security-in-Depth program, illustrating how offensive security research can uncover weaknesses not only in customer environments but in the cloud platform itself. Whether you’re new to OCI or already assessing cloud environments, you’ll leave with a practical understanding of OCI IAM, identity-domain-aware attack path analysis, and an open-source toolkit for exploring OCI security.

People:
    SpeakerBio:  Scott Weston

Originally from Southern California and now based in Minneapolis, Scott has six years of experience in information security. As a Labs Researcher at NetSPI, he builds open-source tools and research material focused on cloud security and penetration testing across AWS, GCP, and OCI environments. His tools include gcpwn, ocinferno, oci-lexer-parser, and ocisigner.

In his spare time, he enjoys being a totally fair and impartial D&D Dungeon Master, and holding out hope that San Diego FC will advance to the MLS finals.

Creator: AI Village
This is a sit down discussion in a more casual conversational format: Topic TBA

People:
    SpeakerBio:  Bruce Schneier, Advisory Board Member at VerifiedVoting.org

Bruce Schneier is an internationally renowned security technologist, called a “security guru” by the Economist. He is the New York Times best-selling author of 14 books – including Rewiring Democracy and A Hacker’s Mind — as well as hundreds of articles, essays, and academic papers. His long-running newsletter and blog, “Schneier on Security,” is one of the most popular sources of cybersecurity news on the internet. Schneier is a Fellow and Lecturer in Public Policy at the Harvard Kennedy School and the Munk School at the University of Toronto. He is a fellow at the Berkman-Klein Center for Internet and Society at Harvard University, a board member of the Electronic Frontier Foundation and AccessNow, and an advisory board member of EPIC and VerifiedVoting.org. He is also the Chief of Security Architecture at Inrupt, Inc.

Creator: Bug Bounty Village

Final words, thanks, and giveaways.

People:
    SpeakerBio:  Bug Bounty Village Staff
No BIO available

Creator: DEF CON Groups
DEF CON 30 sent us home with an idea: our city needed a hacker community.
Three years later, that idea has grown into a 501(c)(3) nonprofit, a thriving monthly meetup, a network that helped us launch a regional conference (wiscon.io), and a long-term vision for a brick-and-mortar hackerspace. Along the way, we’ve learned a lot, made our share of mistakes, and discovered what it really takes to build a sustainable community from the ground up.
This talk is a practical guide to starting and growing a DEF CON Group without burning yourself out in the process. We’ll cover getting approved by DEF CON Groups, incorporating as a nonprofit, raising money from your community, building a board that works well together, writing bylaws that set you up for success, handling Code of Conduct issues, and planning for long-term growth beyond monthly meetups.
We’ll also talk about something that doesn’t get enough attention, building a community as a team. The strongest groups aren’t built by one person doing everything, they’re built by people who share the workload, support one another, and create something that can outlast any single organizer.
Expect honest lessons learned, practical advice, real numbers, and plenty of war stories. We’ll also provide a companion website packed with all the details that you can adapt for your own community.
If you’ve ever thought about starting a DEF CON Group, building a local hacker community, or creating something that brings people together, this talk is for you!

People:
    SpeakerBio:  Ryan Zagrodnik, DC608

Ryan Zagrodnik is a Senior Penetration Tester with over sixteen years of experience in offensive and defensive security roles. He holds OSCP and CISSP certifications and has previously worked on internal red teams and held a U.S. Government security clearance doing offensive security for the Department of Defense and Department of Education. Ryan has presented at CypherCon, SecretCon, DEF CON Hardware Hacking Village, BSides MKE, GRASSr00tz, and Skunks Misery.

SpeakerBio:  Allie Zagrodnik, DC608

Allie Zagrodnik is a cybersecurity professional and community organizer based in Madison, Wisconsin and quickly discovered a passion for the human side of security, community building, and mentorship within the industry.

She is the cofounder of DC608, a Madison-based DEF CON group focused on creating an inclusive and welcoming space for anyone interested in cybersecurity, regardless of skill level or background. She is also the founder of Wisconsin Information Security Conference, a community-driven conference focused on education, networking, and growing the cybersecurity community throughout the Midwest.

Outside of cybersecurity, Allie enjoys spending time with her husband, three dogs, and three cats, while proudly representing Wisconsin through her love of the Green Bay Packers and the local hacker community.

Creator: Policy @ DEF CON

State and local governments are soft targets. Smaller security teams, legacy infrastructure, high-value data, and tight budgets make them attractive to adversaries. Yet most lack basic proactive security programs like Vulnerability Disclosure Programs (VDPs). In January 2025, I was hired as Maryland’s first Adversary Emulation Manager and tasked with standing up a statewide VDP. Coming from DoD cyber operations but new to the bug bounty space, I had to figure out how to build a program that would actually work, not just check a compliance box. The result: Maryland’s VDP launched in October 2025 and has been a success because we fought for true safe harbor, assumed noble intent, and took a researcher-first approach. This talk shares what I learned building the program from scratch. I’ll cover the hard conversations with legal teams who feared “authorizing hacking,” the fights for wide scope instead of narrow carve-outs, and why safe harbor isn’t optional. If you’re a state or local practitioner, you’ll walk away with a replication guide. If you’re a policymaker, you’ll understand why VDPs matter. If you’re a researcher, you’ll see what makes programs succeed or fail.

People:
    SpeakerBio:  01dbae

01dbae has worked in technology since the early 2000s, spanning data center operations, private cloud computing, and information security. After moving into security at a large publisher, he transitioned into government contracting and later served as a DoD civilian supporting the cyber counterintelligence mission. He also spent a few seasons with the Baltimore Ravens.

01dbae later joined the State of Maryland as Adversary Emulation Manager, tasked with standing up the state’s first Vulnerability Disclosure Program — despite having never run a bug bounty program before. That outsider perspective helped him challenge assumptions and push for researcher-first policies over compliance theater. The program launched with strong safe harbor protections and broad scope, and has been running successfully since. There are lessons in that process worth sharing with other state and local governments building proactive security programs

Creator: DEF CON Groups

Healthy DEF CON Groups are built by more than one person. This session focuses on creating a community culture where members are encouraged to participate, contribute, speak, volunteer, and eventually help lead. DC407 will discuss consistency, mentorship, leadership mindset, and practical ways to move people from passive attendance into active involvement. This workshop is for organizers who want to build a group that is welcoming, resilient, and bigger than any single organizer.

People:
    SpeakerBio:  Edna Jonsson, DC407

Edna is one of the DC407 group leaders and enjoys helping others come up in the cyber community. Edna is the volunteer coordinator at BSides Orlando and hosts the Security Chipmunks podcast.

SpeakerBio:  Dustin, DC407

I enjoy helping people. I’m a security enthusiast, if it’s related to security then I’m probably interested. I’ve helped facilitate DC4😃7 for 8+ years.

Creator: OSINT For Good Community

ScamBusters is a student-facing OSINT investigation program that trains students to identify, trace, and report cryptocurrency fraud infrastructure — from pig-butchering scam domains to the wallet networks behind them. This talk shares how the program works, what students have actually found (hundreds of scam domains and wallet clusters tied to organized fraud rings), and how the reporting pipeline connects student research to real-world action.

Links:
    Website – https://www.intelligenceforgood.org/

People:
    SpeakerBio:  Angela Ramos, University of Tampa

Angela Ramos is a University of Tampa cybersecurity lecturer. She leads the Scam Busters student program, coaches Trace Labs Search Party CTFs, and volunteers with US Cyber Games. She holds the GCIH, GSLC, and CEH certifications and spent a decade in DoD cyber operations.

Creator: Lockpick Village

An auto dialer is a device that brute forces the combination to a safe lock. These often start in the thousands of dollars when it’s only a motor that spins. This talk will cover how safe locks work, what goes into building one of these devices, and all the resources (including the source code) for making your own for cheap.

People:
    SpeakerBio:  Jared Dygert
No BIO available

Creator: Packet Hacking Village

PhantomShell is a Linux command & control implant that sniffs traffic of active TCP services to create a bidirectional C2 channel through any open port, using only stateful inbound traffic. It captures packets at the link layer, allowing it to read packets destined for legitimate services. Responses are constructed as raw TCP frames with sequence numbers derived from the original connection, making them difficult to distinguish from the service’s own replies. This effectively turns every open service port into a bind shell, making the implant effectively impossible to firewall so long as any service remains externally accessible.

People:
    SpeakerBio:  Khael Kugler, Lead Offensive Security Engineer, Praetorian Security, Inc.

Khael Kugler is a Lead Security Engineer at Praetorian, where he primarily executes on red team and IoT engagements. Khael also volunteers as a red teamer for multiple CCDC regions, primarily focusing on linux persistence for SECCDC and WRCCDC (if you’re interested in helping, reach out!).

Creator: Hackers.town

People:
    SpeakerBio:  TheGibson
No BIO available

Creator: Recon Village

Building AI-Powered Penetration Testing Bots In this talk, we’ll walk through the core design philosophy behind AI hackbots and the architecture that makes them work: single-purpose vs. multi-stage bots, context engineering for targeted prompting, and tool integration with output parsing workflows. Then we’ll get hands-on. We’ll live-build a functional hackbot for a common offensive task — demonstrating asset discovery, endpoint analysis, or mutation-focused testing (e.g., XSS/SSRF) — and show how context engineering and hallucination mitigation work in practice against real targets. You’ll see where AI genuinely accelerates reconnaissance and web analysis, and where it falls flat if you aren’t careful. We’ll close with lessons on cost optimization, auditability, and integrating hackbots into actual engagements. Who should attend: Pentesters and bug bounty hunters with offensive security experience who want to meaningfully integrate AI into their workflow — not as a novelty, but as reliable tooling. What you’ll walk away with: A clear mental model for when and how to build hackbots, a live demo you can replicate, and a path into the full course where you’ll build seven production-ready bots from asset discovery through mutation testing.

People:
    SpeakerBio:  Jason “jhaddix” Haddix, CEO and “Hacker in Charge” at Arcanum Information Security

Jason Haddix AKA jhaddix is the CEO and “Hacker in Charge” at Arcanum Information Security. Arcanum is a world class assessment and training company.

Jason has had a distinguished 20-year career in cybersecurity previously serving as CISO of FLARE, CISO of Buddobot, CISO of Ubisoft, Head of Trust/Security/Operations at Bugcrowd, Director of Penetration Testing at HP, and Lead Penetration Tester at Redspin. He has also held positions doing mobile penetration testing, network/infrastructure security assessments, and static analysis. Jason is a hacker, bug hunter and currently ranked 57th all-time on Bugcrowd’s bug bounty leaderboards. Currently, he specializes in recon, web application analysis, and emerging technologies. Jason has also authored many talks on offensive security methodology, including speaking at cons such as DEFCON, Bsides, BlackHat, RSA, OWASP, Nullcon, SANS, IANS, BruCon, Toorcon and many more.

Creator: AppSec Village

Building AI-Powered Penetration Testing Bots

People:
    SpeakerBio:  Jason “jhaddix” Haddix, CEO and “Hacker in Charge” at Arcanum Information Security

Jason Haddix AKA jhaddix is the CEO and “Hacker in Charge” at Arcanum Information Security. Arcanum is a world class assessment and training company.

Jason has had a distinguished 20-year career in cybersecurity previously serving as CISO of FLARE, CISO of Buddobot, CISO of Ubisoft, Head of Trust/Security/Operations at Bugcrowd, Director of Penetration Testing at HP, and Lead Penetration Tester at Redspin. He has also held positions doing mobile penetration testing, network/infrastructure security assessments, and static analysis. Jason is a hacker, bug hunter and currently ranked 57th all-time on Bugcrowd’s bug bounty leaderboards. Currently, he specializes in recon, web application analysis, and emerging technologies. Jason has also authored many talks on offensive security methodology, including speaking at cons such as DEFCON, Bsides, BlackHat, RSA, OWASP, Nullcon, SANS, IANS, BruCon, Toorcon and many more.

Creator: Red Team Village
Description: Building AI-Powered Penetration Testing Bots

In this talk, we’ll walk through the core design philosophy behind AI hackbots and the architecture that makes them work: single-purpose vs. multi-stage bots, context engineering for targeted prompting, and tool integration with output parsing workflows.

Then we’ll get hands-on. We’ll live-build a functional hackbot for a common offensive task — demonstrating asset discovery, endpoint analysis, or mutation-focused testing (e.g., XSS/SSRF) — and show how context engineering and hallucination mitigation work in practice against real targets. You’ll see where AI genuinely accelerates reconnaissance and web analysis, and where it falls flat if you aren’t careful. We’ll close with lessons on cost optimization, auditability, and integrating hackbots into actual engagements.

Who should attend: Pentesters and bug bounty hunters with offensive security experience who want to meaningfully integrate AI into their workflow — not as a novelty, but as reliable tooling.

What you’ll walk away with: A clear mental model for when and how to build hackbots, a live demo you can replicate, and a path into the full course where you’ll build seven production-ready bots from asset discovery through mutation testing.

People:
    SpeakerBio:  Jason “jhaddix” Haddix, CEO and “Hacker in Charge” at Arcanum Information Security

Jason Haddix AKA jhaddix is the CEO and “Hacker in Charge” at Arcanum Information Security. Arcanum is a world class assessment and training company.

Jason has had a distinguished 20-year career in cybersecurity previously serving as CISO of FLARE, CISO of Buddobot, CISO of Ubisoft, Head of Trust/Security/Operations at Bugcrowd, Director of Penetration Testing at HP, and Lead Penetration Tester at Redspin. He has also held positions doing mobile penetration testing, network/infrastructure security assessments, and static analysis. Jason is a hacker, bug hunter and currently ranked 57th all-time on Bugcrowd’s bug bounty leaderboards. Currently, he specializes in recon, web application analysis, and emerging technologies. Jason has also authored many talks on offensive security methodology, including speaking at cons such as DEFCON, Bsides, BlackHat, RSA, OWASP, Nullcon, SANS, IANS, BruCon, Toorcon and many more.

SpeakerBio:  Ryan Bonner, Lead Security Engineer at Arcanum Information Security

Ryan Bonner is a Lead Security Engineer at Arcanum Information Security, where he builds AI systems, hacks them, and runs application penetration tests. Off the clock he hunts wide-scope bug bounty programs.

Creator: Blue Team Village

What does it take to build a realistic blue team training platform for thousands of DEF CON attendees? Join the engineers behind Project Obsidian as we discuss the architecture, design decisions, and lessons learned while building a cloud-native incident response and malware analysis CTF. We’ll cover Kubernetes, automation, AI-assisted challenge creation, scalable infrastructure, and the balance between realism and accessibility. Whether you’re interested in CTF design, defensive security, or modern platform engineering, this panel offers an inside look at how a community-driven project comes together.

People:
    SpeakerBio:  Carlo Anez Mazurco

Carlo Anez Mazurco is a cybersecurity instructor, consultant, and community leader with more than 15 years of experience across security operations, threat intelligence, incident response, threat hunting, and detection engineering. He is the Founder of IgniteCyber Academy, a DEF CON Training instructor, and an active contributor to Blue Team Village, where he supports Project Obsidian and develops hands-on blue team content for the cybersecurity community.

Carlo specializes in helping defenders translate attacker tradecraft into practical detection and response techniques while responsibly integrating artificial intelligence into modern security operations. His work focuses on creating realistic labs, CTF challenges, and immersive training environments that prepare students for real-world investigations using enterprise telemetry, cloud technologies, and AI-assisted workflows.

He has delivered training and presentations for conferences, universities, government organizations, and commercial teams, with a passion for mentoring the next generation of cybersecurity professionals. His goal is to make complex security concepts approachable through practical demonstrations, collaborative learning, and hands-on exercises that participants can immediately apply in their own environments.

SpeakerBio:  Chris “dafinga” Maenner, Board Member at BSides Philadelphia

Chris Maenner is the founder of Palmtree Ventures, where he spends his time securing AI systems, Kubernetes, cloud platforms, and developer tooling without getting in the way of shipping software. Over the last 15+ years, he’s worked across startups and Fortune 50 companies building product, platform, and cloud security programs. When he’s not doing his day job, Chris helps build CTFs and security projects for DEF CON’s Blue Team Village and OWASP, including leading engineering efforts around Project Obsidian. He also serves on the board of BSides Philadelphia, contributes to the OWASP Secure Agent Playbook, and regularly speaks about AI security, Kubernetes, cloud-native security, and the lessons learned from actually trying to secure this stuff in production.

SpeakerBio:  Omenscan ~

Some people write books to document and share what they learn. I write software. @Blueteamvillage Director. I do not speak for my employer, their clients, or customers

SpeakerBio:  Paul Goffar

Paul Goffar is the founder of Raven Cybersecurity, doing offensive and defensive security consulting for small and mid-sized businesses. Day to day, he runs incident response, detection engineering, digital forensics, and threat intelligence for an enterprise SOC in the automotive sector.

For the last five years, Paul has worked closely with Blue Team Village as an engineering lead for competitions and content delivery around the US, and does similar CTF and content work with CTF313, a Detroit-based team. He holds several industry certifications including GCIH, GMON, GNFA, CRTP, and paWASP. A father of three, a Metro Detroit native, and a hacker at heart, he continuously finds ways to give back to the community through mentoring and extensive volunteer work.

Creator: Maritime Hacking Village

The Maritime Hacking Village is proud to host a first‑of‑its‑kind conversation with the Principal Cyber Advisors to the Navy and the Air Force. As cyber operations become inseparable from maritime, air, and joint missions, the Navy and Air Force face a shared challenge: how to recruit, train, integrate, and retain the technical talent required for modern conflict.

This session will explore how the services are strengthening operational cyber units. How deeply technical experts can be woven into traditional command structures, and how joint missions can be supported by interoperable training pipelines. We’ll also dig into the realities of competing with the commercial sector for high‑end cyber talent, and the emerging skill sets — from AI engineering to cloud and data‑center operations — that will define future conflict. Expect a little honesty, and maybe one or two jokes about how retention would be easier if the military offered free energy drinks and a hoodie.

People:
    SpeakerBio:  Dr. Nina Kollars, US Naval War College

Dr. Nina Kollars is an Associate Professor at the U.S. Naval War College’s Cyber and Innovation Policy Institute, where she specializes in cyber resilience, emerging technologies, and wargaming efforts focused on Taiwan and the Pacific. Dr. Kollars also serves as co-director of the Maritime Hacking Village, a 501c3 that advances ethical hacking on maritime assets and infrastructure. She is also an executive bourbon steward.

SpeakerBio:  Dr. Wanda T. Jones-Heath, Principal Cyber Advisor for the Department of the Air Force

Dr. Wanda T. Jones-Heath, a member of the Senior Executive Service, is the Principal Cyber Advisor for the Department of the Air Force, comprised of the U.S. Air Force and U.S. Space Force. As the Principal Cyber Advisor for the DAF, her duties include synchronizing, coordinating and overseeing the implementation of the DAF Cyber Strategy and advising the Secretary of the Air Force on all cyber programs. She is responsible for overseeing cyberspace recruitment; resourcing and training of cyber mission forces, as well as assessing their readiness; overseeing acquisition; advocate for cyber investments; cybersecurity supply chain risk management; security of information systems and weapon systems.

SpeakerBio:  Anne Marie Schumann, Principal Cyber Advisor, Department of the Navy

Ms. Anne Marie Schumann serves as the Department of the Navy Principal Cyber Advisor (DON PCA). In this role, she is responsible for advising the Secretary of the Navy, Chief of Naval Operations, and Commandant of the Marine Corps on all cyber matters and implementing the Department of Defense Cyber Strategy within the DON.

Creator: Noob Community

Have you wanted to build a homelab but felt overwhelmed by all the choices? This talk is for you. Building a homelab doesn’t have to be complex or intimidating, I’ll walk you through how easy and simple it can be to set up your first lab and get hands on experience with popular cybersecurity and IT tools.

People:
    SpeakerBio:  Stephen Glombicki
No BIO available

Creator: Red Team Village

Red team operators rely on redirectors to proxy C2 callbacks, but monitoring those redirectors for blue team detection is typically handled by either manual log review or deploying a full SIEM pipeline. Neither scales well during active assessments. NightWatcher is a traffic monitoring agent that analyzes redirector access logs through detection rules and a reasoning engine that identifies attack phase progression (reconnaissance through takedown) rather than issuing disconnected alerts. NightRouter is a decision routing agent that receives NightWatcher’s assessments, recommends a redirector action (drain, reduce weight, quarantine), and presents it to the operator via Slack interactive messages for approval or rejection.

In this tactic, attendees will deploy NightWatcher/NightRouter locally, simulate blue team traffic patterns against redirectors, observe the attack phase reasoning and escalation detection, and execute redirector actions through the Slack human-in-the-loop approval flow. The stack runs in mock mode with no external infrastructure dependencies required.

People:
    SpeakerBio:  Mohamed AbuMuslim

I am a security researcher and offensive security engineer specializing in red teaming, penetration testing, adversarial AI, and product security engineering.

My work focuses on identifying exploitable weaknesses across modern attack surfaces, including web applications, APIs, cloud platforms, enterprise infrastructure, Active Directory, and LLM-enabled systems. I combine offensive operations, applied research, and security engineering to solve complex security problems, validate security posture, and improve how organizations build and assess secure products.

Over the years, I have led and contributed to offensive security capability building in complex environments, including helping establish Microsoft Egypt & Middle East’s first offensive security team and previously helping build PwC’s offensive security capability in Egypt, and served as Manager and Practice Lead at EY leading offensive security engagements. My experience spans startups, mid-sized organizations, and multinational enterprises.

I regularly speak at conferences including Black Hat, DEF CON, BSides, and OWASP Cairo on topics such as AI red teaming, cloud attack simulation, supply-chain risk, log manipulation, and practical offensive tradecraft. I also design and deliver hands-on training, contribute to security education, and support community initiatives through AI Village, BSides Albuquerque, OWASP Cairo, and CyberDose.

SpeakerBio:  Saad Nasir

Saad Nasir is a cybersecurity leader specializing in red teaming, penetration testing, and application security. He currently leads Red Team and Application Security initiatives, overseeing offensive security operations, adversary simulation, and security assessments across enterprise environments.

With more than 10 years of cybersecurity experience, Saad has led offensive security engagements spanning web applications, cloud platforms, internal networks, and Active Directory environments. He is the founder and organizer of Security BSides Albuquerque, helping grow one of New Mexico’s largest community-driven cybersecurity conferences.

Saad is pursuing a PhD in National Security and holds a Master’s degree in Cybersecurity, along with multiple industry certifications, including OSCP and CISM. He is passionate about sharing practical skills, helping defenders understand the mindset of attackers, and advancing the cybersecurity community through mentoring, conference speaking, and hands-on training.

Creator: Mobile Hacking Community

In this talk, the Just Mobile Security team will talk about the KYC implementation, how to bypass them, and some 0 days for vendors.

People:
    SpeakerBio:  Juan Urbano Stordeur, CEO and Founder at Just Mobile Security
No BIO available
SpeakerBio:  Juan Martinez Blanco, Mobile Security Penetration Tester at Just Mobile Security
No BIO available

Creator: Bug Bounty Village

Cache poisoning research has long centred on unkeyed-input injection. The next cache poisoning vulnerability may not come from unkeyed input, but from what is mistakenly included in the cache key itself. This talk explores a lesser-known attack path: cache key injection caused by subtle cache misconfigurations that allow attacker-controlled values to influence keyed cache components. By triggering cache key injection, attackers can perform cache key collisions, leading to cache poisoning and impacts such as CPDoS, cache deception, and stored XSS through poisoned cache keys. I’ll show how these collisions can be identified, how they can be exploited to achieve impact, and the mitigations that can be implemented to prevent cache key injection.Cache poisoning research has long centred on unkeyed-input injection. The next cache poisoning vulnerability may not come from unkeyed input, but from what is mistakenly included in the cache key itself. This talk explores a lesser-known attack path: cache key injection caused by subtle cache misconfigurations that allow attacker-controlled values to influence keyed cache components. By triggering cache key injection, attackers can perform cache key collisions, leading to cache poisoning and impacts such as CPDoS, cache deception, and stored XSS through poisoned cache keys. I’ll show how these collisions can be identified, how they can be exploited to achieve impact, and the mitigations that can be implemented to prevent cache key injection.

People:
    SpeakerBio:  Alex Brumen, YesWeHack

Alex Brumen (aka “Brumens”) is a Security Research Enablement Analyst at YesWeHack, where his ethical hacking work focuses primarily on web applications. Outside of work, he is also a bug bounty hunter and developer.

Creator: Payment Village

Join us as we announce the winners of the Contest!

Creator: Game Hacking Village

Super Smash Bros Melee for the Nintendo GameCube is one of the oldest active fighting game communities, with a storied 25 year history.

Would you believe that people try to cheat at it?!

I am the maintainer of the SLP Enforcer tool, and help with cheating investigations for the community. I’m going to share some stories of people trying to cheat and getting caught! Along the way I’ll describe all the ways one could try to cheat at Melee, and how our detection tools work.

Links:
    altf4.github.io/enforcer/ – https://altf4.github.io/enforcer/

People:
    SpeakerBio:  AltF4

Dan “AltF4” Petro is a Principal Security Engineer at Bishop Fox R&D. As a longtime pentester at Bishop Fox, Dan’s presented public research on everything from hacking into smart safes, compromising the power grid, breaking into secure facilities, and cheating at online video games. There was even that time he found a vulnerability in (nearly) every IoT device. But Dan’s favorite research project is always whichever is next.

Creator: Red Team Village

AI/ML services are proliferating across enterprise networks without security review. Developers deploy Ollama instances locally, data science teams stand up Jupyter notebooks and MLflow registries without authentication, platform engineers run Ray clusters and LiteLLM API gateways on default configurations, and MCP tool servers expose file system access and code execution to the network. These services rarely appear in traditional vulnerability scans and are seldom included in asset inventories.

To measure the scope of this problem, I built a 5-VM benchmark lab simulating four independent teams deploying 19 AI/ML endpoints with 122 planted findings. The lab includes a manifest-driven scoring harness that enables objective detection measurement across three validation modes: standard (substring matching), strict (host attribution and source module accuracy), and contract (workflow metadata and evidence chain verification).

The tool is aipostex, an open-source single-binary Go framework designed for the full AI infrastructure attack lifecycle. It performs network discovery across 20+ AI service families, executes 123 vulnerability templates, and provides 17 dedicated exploit modules covering Ollama, Jupyter, MCP servers, LiteLLM gateways, MLflow, Ray, vector databases, and inference endpoints. MCP coverage includes schema poisoning, environment variable extraction, and CVE-specific checks for DNS rebinding and remote code execution through tool servers. A separate model-scan capability identifies deserialization risk in pickle, PyTorch, TensorFlow, and ONNX model files.

The live demonstration walks through a credential chain attack against the benchmark lab. The attack begins with Ray cluster enumeration to harvest API keys from job environment variables, pivots into MLflow where experiment run parameters and tags contain embedded secrets and cloud credentials, then chains those tokens to authenticate against a HuggingFace TGI inference endpoint. Each hop crosses a different team’s infrastructure, discovered and linked by the tool’s credential chain-loading engine.

The session also covers the safety and operational controls that make this suitable for production engagements: explicit –force-exploit gating on mutating actions, –mode full requirement for exploit templates, proof-strength classification on every finding (reachable, read-confirmed, execution-confirmed), and OPSEC features including User-Agent rotation, TLS fingerprint randomization, and timing jitter.

Open-source release coinciding with presentation.

Does your workshop come with a tactic?

Yes. The tactic is “Hands-On: Credential Chain Exploitation Across Shadow AI Infrastructure.”

Each group of about five attendees gets their own isolated cloud-hosted lab instance, a full 5-VM environment spun up on AWS and accessible only through a VPN tunnel. You SSH into your group’s attack box and work through a guided attack chain against the lab’s 19 AI/ML endpoints.

First, you scan the /24 and discover what’s running across four target hosts. The tool fingerprints each endpoint and gives you structured next steps, so even if you’ve never touched MLflow or Ray before, you know what to run next.

Then you follow the credential chain. Enumerate a Ray cluster and pull API keys from job environment variables. Take those into MLflow and extract secrets from experiment run parameters and tags. Chain the tokens forward to authenticate against a HuggingFace TGI inference endpoint. Three hops, three teams’ infrastructure, all connected.

At the end, you run the scoring harness against your results and see how you did. Detection rate, missed findings, proof-strength breakdown, all compared against the 122-finding answer key.

Groups that finish the guided chain early can explore additional attack surfaces across the lab’s remaining endpoints, including MCP schema poisoning, Jupyter cell secret mining, and vector database injection. The scoring harness covers all 122 findings, not just the guided path, so there is plenty of room to improve your score.

Each wave runs about 50 minutes with a 10-minute reset between waves. I’ll run two waves. All you need is a laptop with an SSH client and a WireGuard client. No prior AI/ML experience required. Everything you use during the tactic, the tool binary, lab docs, and scoring harness, is yours to take home and run on your own infrastructure.

https://professor-moody.github.io/aipostex/

https://professor-moody.github.io/aipostex-lab/

People:
    SpeakerBio:  Nathan Keys

Nathan is a security researcher focused on ML supply-chain security. Their research spans information hiding in model artifacts, data poisoning of retrieval pipelines, and post-exploitation of AI infrastructure. Nathan is currently a principal penetration tester in the financial sector. This is their first DEF CON talk. Outside of his passion for research, Nathan loves to touch grass, read all manner of scientific study or journal, and listen to old school southern rap (think UGK). Nathan has changed careers more than once, from winemaking to restaurants to professional hacking now for the last seven years, and he loves a good story and a good conversation.

Creator: Blacks In Cyber Village

Join B.I.C. Village for ChronoRoot: Time-Traveling Through Kernel Trust Boundaries, a welcoming session focused on kernel trust boundaries, system internals, and how attackers challenge low-level security assumptions. Come ready to learn, ask questions, and connect with the community.

People:
    SpeakerBio:  Ahmeen Muhammad, Cybersecurity Professional / Community Speaker

Ahmeen C. Muhammad is a Senior Penetration Testing Consultant and Army cyber veteran with experience conducting offensive security operations across enterprise and government environments. His interests include red teaming, reverse engineering, Malware Development, endpoint security evasion, and AI security testing. When he’s not on an engagement, Ahmeen spends his time researching emerging attack techniques, developing offensive tooling, and sharing knowledge with the cybersecurity community. He is passionate about helping organizations better understand their security posture through realistic adversary emulation and hands-on offensive security research.

SpeakerBio:  Sydney Johns, AI/ML Engineer, Virginia Tech Researcher

Sydney Johns is an Artificial Intelligence Researcher at GitHub and a Ph.D. student in Computer Science at Virginia Tech. Her research interests include artificial intelligence, machine learning, reverse engineering, and ethical hacking. She holds a Bachelor’s degree in Electrical Engineering and a Master’s degree in Cybersecurity, along with CompTIA Security+ and EC-Council Certified Ethical Hacker certifications. Sydney is passionate about STEM outreach, mentorship, and creating pathways for more people to enter technology and cybersecurity. In her spare time, she enjoys playing Overcooked and Animal Crossing, painting, and shopping.

Creator: Cloud Village

Creator: Recon Village

Cl0p has executed at least nine major exploitation campaigns since 2020, targeting file transfer and edge devices from Accellion to MOVEit to Centrestack. Each campaign looks different on the surface, different CVEs, different victims, different tooling. But their infrastructure tells a different story.

This talk presents the results of a multi-year infrastructure reconnaissance effort tracking Cl0p’s hosting, ASN usage, and operational patterns across all known campaigns. By mining passive DNS data, network telemetry, and hosting records, I mapped the infrastructure behind each campaign and identified patterns the group can’t seem to shake — including a favorite bulletproof hosting provider used across four separate campaigns, a finding that roughly one-third of their ASNs get recycled, and pre-attack reconnaissance probing observed as far as two years before exploitation.

People:
    SpeakerBio:  Eli Woodward, Senior Threat Intelligence Advisor with Team Cymru

Eli Woodward is a cyber threat intelligence advisor with Team Cymru. He’s worked in a variety of industries including financial services and government, and has seen the range of organizations from extremely well-resourced and capable to the shoestring budget. This has given him unique insights into CTI at all levels of complexity, maturity, and resources. He holds a master’s degree in Intelligence and Security Studies and also plays bagpipes competitively.

Creator: Lockpick Village

With pins that both slide up and down and rotate, Medeco locks have been one of the most common high security locks in the US for over 50 years. Come learn exactly how they work, and how to open them with standard picks.

People:
    SpeakerBio:  Max A
No BIO available

Creator: OSINT For Good Community

AI this, AI that, and of course even in the world of OSINT AI has been getting integrated. But this talk focuses on a different AI: Actual Intelligence. Modern investigators have access to more tools, automation, and artificial intelligence than ever before, yet many investigations still suffer from poor assumptions, weak validation, and information overload. Long before automated platforms and AI assistants existed, investigators relied on analytical thinking, source evaluation, and structured methodology to turn information into intelligence. This session introduces the ACTUAL framework. Through practical examples and real-world investigative scenarios, attendees will learn how to validate findings with confidence and conduct effective OSINT investigations even when specialized tools aren’t available. Tools help collect information, but ACTUALly intelligence comes from the investigator.

People:
    SpeakerBio:  Bianca C. Ionescu/reconcerto

Bianca Ionescu is a CyberCorps SFS scholar pursuing a master’s degree in cybersecurity at UNLV. She’s served as a leader within cybersecurity student organizations, promoting growth, inclusion, and professional development. Her interests include open-source intelligence and mentoring new learners entering the field. Offline, she enjoys strength training and playing the viola. She’s motivated by community building, continuous learning, and the challenge of solving complex security problems that inspire her growth and resilience every day.

Creator: Aerospace Village

Cybersecurity threats are rapidly becoming a first-order safety concern in critical aviation systems. Increasing in-flight connectivity, satellite communications, and networked avionics have eroded the assumption of airborne isolation. At the same time, AI-generated attacks are lowering both the barrier to entry and the time-to-exploit, enabling faster, more adaptive threat behavior. In contrast, patching and remediation time in aviation systems remains long due to rigorous certification and deployment constraints. This growing asymmetry further elevates cybersecurity risk into a direct safety concern. In this environment, a security risk is a safety risk. This raises a critical question: not whether cybersecurity is needed in aircraft, but whether cybersecurity tools can be certified for safety-critical airborne systems.

This talk focuses on the challenge of bridging that gap by asking: can we generate the right verification artifacts to make existing cybersecurity tools certifiable, without modifying their core functionality? We explore why cybersecurity tools are rarely introduced into DAL-A airborne systems in their native form, focusing on the mismatch between operational security outputs and certification evidence requirements. The core problem is not only deterministic behavior, but the lack of structured, complete, and traceable artifacts to support certification arguments.

People:
    SpeakerBio:  Katie Fejer
No BIO available

Creator: Bug Bounty Village

Bug bounty hunters often encounter unprotected URL inputs and dismiss them as low-impact findings. This presentation explores real-world Microsoft enterprise application vulnerabilities where custom URI handlers led to remote code execution. Attendees will learn how pivoting from web applications into desktop software can reveal impactful attack paths hidden behind seemingly mundane bugs.

People:
    SpeakerBio:  Tobias Diehl

Tobias Diehl is a Senior Offensive Security Engineer, security researcher, and Microsoft 2025 Most Valuable Researcher (MVR) specializing in offensive security, enterprise AI, bug bounty research, and adversary emulation. His work focuses on identifying overlooked attack paths where web applications, AI systems, and desktop software intersect, helping organizations understand how seemingly low-risk vulnerabilities can become high-impact security issues. Tobias leads offensive security assessments, conducts purple team exercises, and performs research into emerging attack techniques affecting enterprise environments. His work has resulted in multiple security findings impacting Microsoft technologies, including Power Platform, CoPilot and other AI-driven applications. A returning DEF CON speaker, Tobias is passionate about sharing practical research that helps both security researchers and defenders better understand modern attack chains. His presentations emphasize real-world case studies, responsible disclosure, and actionable defensive guidance for securing the next generation of AI-enabled enterprise systems.

Creator: Physical Security Village

Traditional RFID badge cloning methods require you to be within 3 feet of your target. So how can you conduct a physical penetration test and clone a badge without interacting with a person? Companies have increasingly adopted a hybrid work environment, allowing employees to work remotely, which has decreased the amount of foot traffic in and out of a building at any given time. This session discusses two accessible, entry-level hardware designs you can build in a day and deploy in the field, along with the tried-and-true social engineering techniques that can increase your chances of remotely cloning an RFID badge.

Langston and Dan discuss their Red Team adventures using implant devices and their Flipper Zero workflow. As a bonus the two will present a new python script to help you decode your badge data faster. This presentation is supplemented with files and instructions that are available for download in order to build your own standalone gooseneck reader, wall implant and clipboard cloning devices!

Links:
    Github – https://github.com/sh0cksec

People:
    SpeakerBio:  Langston Clement, Security Risk Advisors (SRA)

Langston grew up reading stories about the 90’s hacker escapades, and after years of observing the scene, he jumped into the offensive security field and never looked back. He is currently a Senior Purple Team Advisor for Security Risk Advisors (SRA). With over fifteen (15) years of public and private sector experience in cybersecurity and ethical hacking, he aims to provide organizations with valuable and actionable information to help improve their security posture. Langston’s specializations focus on modern-day social engineering techniques, wireless and RFID attacks, red teaming, physical penetration testing and purple team engagements.

SpeakerBio:  Dan Goga, Principal Consultant at NRI North America

Dan Goga serves as a Principal Consultant with NRI focused on conducting penetration testing and red team assessments. Dan Goga has nine years of information security experience in the public, private, and academic sectors. Dan has extensive knowledge and experience with RFID hacking, phishing techniques, social engineering techniques, and penetration testing.

Creator: Cloud Village

People:
    SpeakerBio:  Jayesh Singh Chauhan
No BIO available

Creator: Blue Team Village

Cloud breaches don’t look like Hollywood hacking; they look like a leaked API key and a bot finding it before you’ve finished your coffee. This introduction to cloud forensics talk walks through a real AWS attack path end-to-end: recon, initial access, privilege escalation, exfiltration, and impact. We’ll cover the misconfigurations that actually get exploited in the wild, how to contain them, and close with a demo: a deliberately compromised AWS account, investigated using AI to turn raw CloudTrail logs into a clean incident timeline in minutes.

People:
    SpeakerBio:  Cassandra (muteki) Young

As a Principal Consultant on [REDACTED]’s Cloud Technical Advisory team, muteki analyzes the organizational security posture of Azure, GCP and Oracle Cloud environments, and leads the development of data collection and analysis tooling. Additionally, she assists the Incident Response team as a technical GCP and OCI SME, and supports strategic advisory and technical tabletop exercises across multiple cloud platforms. In addition to her decade of IT and then cloud security consulting experience, muteki also holds a Master’s in Computer Science and is a director of Blue Team Village, a nonprofit organization bringing free Blue Team content to the community.

SpeakerBio:  Dimple Gajra

Specializing in Digital Forensics, Incident Response (DFIR), and privacy engineering, Dimple Gajra (aka LocalHost) brings technical expertise to enterprise defense and data protection. Her professional journey includes engineering and response roles at major technology enterprises like IBM and Amazon’s Security Incident Response Team (SIRT), where she has actively defended critical infrastructure, mitigated high-severity ransomware incidents, and engineered solutions to safeguard data privacy and automate secure detection pipelines. Driven by a hands-on, analytical approach to uncovering adversary tradecraft, she focuses on building architectural resilience, protecting sensitive user data, and translating complex system artifacts into actionable defensive strategies.

Creator: Cloud Village

Modern engineering teams don’t live in one cloud. A workload in Google Kubernetes Engine needs to read from an S3 bucket. An EC2 instance needs to fetch a secret from Azure Key Vault. A single GitHub Actions job has to deploy to AWS, log telemetry to GCP, and trigger a workflow in Azure — all in the same run. The default way teams make this work is the same way they have for a decade: paste a long-lived credential into a secret and hope nobody finds it.

This is the credential sprawl problem at the heart of every multi-cloud breach. Static AWS access keys, GCP service account JSON files, Azure client secrets — the most reliably exploited foothold in cloud environments, sitting in env vars, secret managers, and committed git history across every blue team’s environment.

Why this was hard before cloud-auth? In theory, every major cloud already supports a way to skip the static credential entirely. A GKE pod can assume an AWS IAM role directly — no AWS keys anywhere. An EC2 instance can call Azure with no service principal on it. A Kubernetes workload can hit GCP APIs with no service account JSON file. In practice, each cloud has its own model, API, and failure modes. Setting up just one of these connections means learning a new mental model, authoring trust documents with provider-specific subject formats, and validating it all by hand. Doing it across clouds means doing that work three or four times over, with no consistent way to set it up, validate it, or clean it up. That friction is why cloud-auth exists. The secure path is already supported by the clouds — defenders just need tooling that makes it as easy as pasting an access key, or easier. So most teams stay on static keys.

cloud-auth: cross-cloud workload identity for defenders cloud-auth is an open-source CLI and library that gives blue teams one consistent way to set up, validate, and safely retire cross-cloud workload identity. One command stands up the trust between any two of: AWS, GCP, Azure, HashiCorp Vault, Cloudflare, GitHub Actions, and Kubernetes. The same command validates it, previews changes before applying, and tears it down without touching anything it didn’t create.

Live demo GKE pod → AWS S3, no AWS keys. A GKE pod reads from S3 using a short-lived AWS role. EC2 → Azure Key Vault, no service principal. An EC2 instance fetches a secret with no Azure client secret on it. GitHub Actions → AWS, GCP, and Azure in one workflow. A single CI job deploys across three clouds. Catching a misconfigured trust. The validator flags an overly permissive trust before it ships. Safe teardown. Dry-run a multi-cloud cleanup, then execute. cloud-auth only deletes what it created.

What attendees leave with? A working open-source tool, a concrete pattern for cross-cloud workload identity, and a validation framework to drop into existing CI/CD. Repository: https://github.com/anirudhbiyani/cloud-auth (LGPL-3.0)

People:
    SpeakerBio:  Aniruddha Biyani

Aniruddha “AB” Biyani leads the Security and Compliance program at Prophecy, where he focuses on building security as a business enabler for high-growth cloud and AI platforms. His work centers on scaling information security programs, strengthening cloud and identity security, enabling secure product development, and aligning security strategy with customer trust, compliance, and engineering velocity. He has previously held security roles at Mandiant, Securonix, CRED, and Khoros.

AB’s career spans cloud security leadership across startups and cybersecurity vendors, with deep experience in defensive cloud engineering, identity and access management, secure developer enablement, and operationalising secure-by-default practices in fast-moving organizations. He is known for translating complex security challenges into practical programs, tools, and frameworks that help engineering teams move quickly while reducing risk.

AB has presented at global security conferences including Black Hat Arsenal, SANS CloudSecNext, and MCPDev BLR, on workload identity, multi-cloud attack surface discovery, credential lifecycle management, and secure cloud adoption at scale. He is also an active open-source contributor and community builder, maintaining cloud security tools and secure developer training resources. He previously served for three years as a core DEF CON Cloud Village CTF volunteer, leading contest design and build efforts.

Creator: Red Team Village

Traditional command and control infrastructure faces constant detection pressure from network defenders. This research presents two novel proof-of-concept frameworks that demonstrate how seemingly benign cloud services can be weaponized for covert operations. MeetC2 leverages Google Calendar API to transform calendar events into a bidirectional C2 channel, while XRayC2 repurposes AWS X-Ray’s distributed tracing service for command execution and data exfiltration.

Both frameworks exploit fundamental assumptions about trusted services: calendar synchronization traffic appears as routine business operations, while X-Ray traces blend seamlessly with legitimate application monitoring. Our implementations showcase practical techniques including command encoding in calendar event metadata, response embedding in trace annotations, and beacon patterns that mimic normal service behavior. The frameworks support cross-platform implants with minimal footprint, requiring only API credentials rather than traditional malware persistence.

I will demonstrate how these channels bypass traditional security controls, as encrypted HTTPS traffic to Google and AWS endpoints rarely triggers alerts. The research reveals critical blind spots in cloud security monitoring and provides actionable detection strategies, including API usage anomalies, unusual trace patterns, and calendar event behavioral analysis. This work aims to help defenders understand emerging cloud-native threats and implement appropriate monitoring for their cloud environments.

As organizations increasingly adopt cloud services, attackers have evolved their tactics to abuse these trusted platforms for malicious purposes. Cloud service APIs present an attractive option for establishing command and control infrastructure because they offer high availability, encrypted communications by default, and traffic that blends with legitimate business operations. This research explores how adversaries can weaponize standard cloud APIs to create sophisticated C2 channels that circumvent traditional security controls.

This research presents two functional C2 frameworks for initial access that abuse legitimate cloud APIs: MeetC2 leveraging Google Calendar for command and control, and XRayC2 weaponizing AWS X-Ray distributed tracing.

People:
    SpeakerBio:  Dhiraj Mishra

An active speaker who has discovered multiple zero-days in modern web browsers and an open-source contributor. He is a trainer at Blackhat, BruCON, 44CON and presented in conferences such as Ekoparty, NorthSec, Hacktivity, PHDays, Hack in Paris & HITB. In his free time, he blogs at www.inputzero.io/www.fuzzing.at and tweets on @RandomDhiraj.

Creator: Blue Team Village

As organizations scale across a fragmented mix of SaaS, PaaS, and hyperscale environments, the enterprise attack surface is no longer fully under their control. Today’s threat actors aren’t just targeting the cloud, they are weaponizing the cloud supply chain and exploiting these new dependencies. Bringing together diverse practitioner perspectives, this panel addresses the hard operational questions of modern cloud security.

Our experts will break down the structural differences in evaluating risk across various cloud tiers, share proactive blueprints for hardening application supply chains, and debate the unique pitfalls of executing incident response when a critical SaaS provider or cloud asset is compromised.

People:
    SpeakerBio:  Cassandra (muteki) Young

As a Principal Consultant on [REDACTED]’s Cloud Technical Advisory team, muteki analyzes the organizational security posture of Azure, GCP and Oracle Cloud environments, and leads the development of data collection and analysis tooling. Additionally, she assists the Incident Response team as a technical GCP and OCI SME, and supports strategic advisory and technical tabletop exercises across multiple cloud platforms. In addition to her decade of IT and then cloud security consulting experience, muteki also holds a Master’s in Computer Science and is a director of Blue Team Village, a nonprofit organization bringing free Blue Team content to the community.

SpeakerBio:  Christian Nicholson

Christian is a recognized technology leader specializing in cloud architectures and secure-by-design implementation. Maintaining deep technical expertise, Christian bridges the gap between IT, business teams, and users. A global speaker, advisor, and organizer for major IT and security conferences and global events, Christian also contributes to international policy through United Nations (UN) and global policy dialogues to shape secure digital ecosystems.

SpeakerBio:  David Collins

Dave works to secure modern infrastructure and applications, with a focus on cloud-deployed systems.

SpeakerBio:  Kyle Dickinson

Kyle D is a professional overthinker who’s spent years securing complex systems at scale without killing the fun. A former Threat Detection and Response specialist at {Cloud Provider Here), he now lives at the intersection of cloud, gaming, and security.

SpeakerBio:  Ricky (0xpsilocyber) Banda

Hi!

I am a incident responder with over 15 years of experience with a focus on incident management. I have worked at orgs from USAF, to Amazon, to Google, and have responded to a wide array of threats across the industry. Feel free to say hi after the panel, always happy to chat, discuss, share, and meet new folks!

Creator: Red Team Village

Component Object Model (COM) is one of the most pervasive yet overlooked subsystems in Windows. Thousands of applications rely on COM objects for functionality, but the underlying registration mechanism creates opportunities for attackers to abuse the Windows registry to achieve stealthy persistence and code execution. Despite being used in real-world intrusions, COM hijacking remains under-documented and poorly understood by many defenders.

This talk explores the internals of COM registration and activation and demonstrates how attackers abuse registry-based class registrations, InprocServer32 entries, and per-user overrides to hijack legitimate COM objects. We will analyze common hijacking patterns, explain why many of them evade traditional detection logic, and highlight the operational advantages they offer during post-exploitation and persistence phases.

To help security professionals systematically identify these opportunities, this presentation introduces COM-Hunter, a research tool designed to enumerate and analyze hijackable COM objects across Windows systems. COM-Hunter maps CLSID registrations, identifies missing or user-overridable components, and highlights potential hijacking vectors that can be leveraged during red team engagements or security research.

Through practical demonstrations, we will show how COM-Hunter can be used to uncover previously overlooked hijack paths and how red teamers can turn these findings into reliable persistence mechanisms. The talk also discusses defensive considerations, including detection strategies, telemetry sources, and ways organizations can reduce the attack surface created by vulnerable COM registrations.

Attendees will leave with a deeper understanding of COM hijacking internals, practical offensive techniques, and a methodology for discovering new hijack opportunities in modern Windows environments.

People:
    SpeakerBio:  Nikos “nickvourd” Vourdas, EY

Nikos Vourdas, also known as nickvourd or NCV, is a Senior Offensive Security Consultant based in the US. With over five years of professional experience, he has actively participated in various global Tiber-EU and iCAST Red Teaming engagements. Nikos has conducted full Red Teaming operations to major clients across retail, banking, shipping, construction industries. He holds OSCE3, OSCP, OSWP, CRTL, CRTO and OASP certifications. Also, he has previously presented at DEF CON, DevSecCon, and various BSides events around the world. Nikos loves contributing to open-source projects and always starts his day at 05:00 AM with a refreshing jog while listening to French rap music.

Creator: Red Team Village

This hands-on workshop provides a unified view of C2 fundamentals for both offensive and defensive practitioners. Using the open-source Mythic framework, participants will deploy agents, handle callbacks, execute tasking with commentary on opsec and payload design considerations.

The session will also cover basic C2 infrastructure design including redirectors/domain fronting and an overview of Mythic feature sets. Students should leave armed with practical introductory experience operating Mythic for Red Team engagements.

The requested 2.5 hour timeslot should allow for at least two complete runs through the presentation and hands-on workshops as well as time for Q&A and subsequent “turnover” on the room for a new wave of attendees.

Participants must have the following equipment:

• A laptop capable of running Docker containers and/or macOS/Windows/Linux VMs. It is highly recommended that students do not use corporate assets for this workshop as they are likely to trigger AV/EDR when working with the example (default) payloads.

• If students wish to host Mythic themselves an x86 host/VM is required, however this will not be part of the workshop. We will provide a limited-access, cloud-hosted Mythic C2 server environment for students to connect to.

People:
    SpeakerBio:  Logan MacLaren

Logan is the lead Offensive Security engineer at Huntress where he is responsible for planning and executing red team operations as well as bolstering incident response capability through purple team exercises. He has been a long time enthusiast in the security space, building a career spanning big data analytics, bug bounty, and offensive security.

Outside of his day job, Logan can often be found building and participating in CTF challenges, bug hunting in open source software, or learning new skills at conferences across the continent. He has had the honour of speaking at several DEFCON villages, NorthSec conferences, as well as multiple BSides and OWASP Ottawa events.

Creator: Red Team Village

GitHub is where identity, automation, and production changes all meet, which makes it a great target. Once an attacker gets in, the distance between read access and shipping malicious code to production is often a lot shorter than teams expect.

This talk covers realistic attack paths into GitHub organizations. We start with initial access: device-code phishing and abuse of trusted GitHub Apps like the GitHub CLI. Then we follow the credentials, from long-lived Personal Access Tokens sitting on developer machines to short-lived secrets like GITHUB_TOKEN that leak through logs and artifacts, and show how both enable lateral movement and privilege escalation.

Then we get into a technique we call secret stomping. Anyone with write access to a repo can overwrite an Actions secret, including protected environment secrets reviewers assume are safe. Leak the real value first, then overwrite the secret with a payload that keeps the original value intact, and the pipeline stays green while your code runs on the runner. Using the org, repo, and environment scope hierarchy, you can shadow a secret in one repo without touching others, then delete it and leave nothing behind at the org level.

We close on the defensive side with detection strategies and response playbooks built around the signals that matter, plus how to get coverage into the places GitHub is hardest to watch. Blue teams leave with a checklist for locking down identities, tokens, integrations, and workflows. Red teams leave with a map of where GitHub controls break in practice.

After the talk we’re running a hands-on workshop in a private GitHub org built for this, focused on secret stomping. You’ll leak a secret, stomp it with a payload that keeps the pipeline green, and use scope shadowing to exploit your way onto a sensitive runner without tripping the obvious alarms. Bring a laptop.

People:
    SpeakerBio:  Andrew Buchanan

Andrew Buchanan is a Senior Red Team Operator with over six years of offensive security experience spanning adversary simulation, penetration testing, and real-world attack execution.

Andrew has spent years conducting advanced red team engagements and security assessments across highly complex enterprise environments at one of Canada’s largest financial institutions.

Andrew specializes in initial access and social engineering, having designed and delivered numerous campaigns that closely mirror real-world threat actor tradecraft. His work spans offensive operations across on-premises, cloud, and hybrid environments, with a particular focus on cloud attack surfaces, execution chains, and targeted phishing and pretexting campaigns.

SpeakerBio:  Max CM

Max leads offensive security at Figment and brings over a decade of experience spanning national security, security research, and blockchain security. He has published multiple CVEs and conducts research focused on CI/CD pipeline security, threat modeling, secure key management, and practical security controls for high-risk systems.

Creator: Nix Vegas Community

Reverse engineering binaries is slow, detailed work, but it’s one of the few ways we have to verify what some software actually does. Machine learning might help improve some of those challenges, but only if you have high-quality, real-world ground truth to train and evaluate on. At MIT Lincoln Laboratory, we have built a large dataset for machine learning on x86_64 binaries using nixpkgs, taking advantage of Nix’s reproducibility, package coverage, and instrumentable build environments. We compiled tens of thousands of C, C++, Rust, and Go packages, often multiple versions of each, with consistent compiler flags, captured source files, and debug information. From these, we extracted roughly 50 million functions and aligned with their source code and short descriptions. Possible applications include fine-tuning large language models for various binary tasks or training custom, highly efficient models for binary code understanding. I’ll walk through what it takes to wrangle nixpkgs in an HPC environment at this scale and show how open build tooling like Nix can help us to better understand and analyze the software that we all depend on.

Links:
    cfp.nix.vegas/2026/schedule/ – https://cfp.nix.vegas/2026/schedule/

People:
    SpeakerBio:  Chris Connelly, MIT Lincoln Laboratory

Chris Connelly is a Senior Technical Staff member at MIT Lincoln Laboratory in the Cyber Security and Information Sciences Division. Since joining the Laboratory in 2002, he has contributed to or led research, development, and evaluation programs in several areas, including malware analysis, vulnerability discovery, systems analysis, test automation technologies, large-scale distributed systems, and software architecture.

Creator: Nix Vegas Community

MicroVMs make it easy to give every service, project, or person dedicated NixOS machines. We will use microvm.nix to build dev boxes, game servers, and sensitive services on hardware we control. Because each guest is simply a NixOS configuration, it can be reused, extended, rebuilt, and shared.

We will compare MicroVM isolation with other options, explain its limitations, and show how to compose useful machines from ordinary NixOS modules for yourself and those around you. We will finish with a look beyond a single host: how to preserve and back up these machines, then place them across hardware you own.

People:
    SpeakerBio:  Alex Decious

I work on build systems and CI infrastructure for Shopify’s monorepo using Nix, and run my own infrastructure as a systems lab for compute-heavy builds, distributed systems experiments, local LLMs, and hosting for myself and others.

Creator: Payment Village

ATMs are the ultimate high-stakes target, often holding upwards of $400,000 in a single enclosure. While the global financial industry relies on a narrow pool of manufacturers, the software supply chain securing these “vaults” remains an under-examined attack surface. Following my disclosure of 6 code execution vulnerabilities affecting Diebold Nixdorf at DEF CON 32, this research dives deeper into the foundational security layer: CryptWare CryptoPro Secure Disk for BitLocker.

CryptoPro acts as a proprietary security wrapper, adding pre-boot authorization, custom TPM protections, and an obfuscated encryption layer to the standard BitLocker architecture. In this session, I will highlight the nuances of the CryptoPro architecture, including, secret storage through unallocated disk space, TPM sealing logic, and the layered crypto architecture used to secure system secrets. I will demonstrate how these deficiencies provide a path for code execution and full compromise of the Windows BitLocker encryption keys.

In addition to the technical walk through, I will release ragavan, a custom exploitation toolkit designed to automate secret extraction, decryption, TPM unsealing, and compromise of a CryptoPro-protected platform.

People:
    SpeakerBio:  Matt Burch, Principal Security Researcher at Atredis Partners

Matt Burch is a Principal Security Researcher at Atredis Partners with 20 years of experience breaking things that aren’t supposed to break. From the embedded components of ATM platforms to complex SCADA/OT environments, Matt specializes in identifying critical flaws in hardened, enterprise-class systems. He is best known for his research into ATM disk encryption and Mobile Device Management (MDM) security, some of which has been highlighted in Wired Magazine and presented at DEF CON 32.

Matt is a reverse engineer and tool developer who has authored and contributed to various public projects. His career spans roles as an offensive security lead, expert witness, and technical advisor. Matt’s current research is a deep-dive into the ATM software supply chain, specifically targeting the subversion of TPM-backed roots of trust and the analysis of custom cryptographic primitives.

Creator: Middle Easterns & Africans in Cyber Security (MEACS)

Learn to use ML to condense volumes by 80% without data loss by grouping related records and deduping their redundant field values. Learn how to use ML to to auto-normalize records to a unified schema and enrich them with their most relevant tags and labels across objects, assets, entities, frameworks and controls. Use ML to correlate situationally relevant records, similar root causes and link causal sequences and behavior. Use AI to further troubleshoot, detect and remediate with rich-context.

People:
    SpeakerBio:  Ezz Tahoun

Ezz Tahoun is an award-winning cybersecurity data scientist recognized globally for his innovations in applying AI to security operations.

He has keynoted, trained & presented at BlackHat US, Sector, MEA, Asia & EU, DEFCON, SANS Summits, all the top Bsides, Securityweek ICS Conference and GISEC among many others.

His groundbreaking work earned him a gold edison award and accolades from Yale, Princeton, Northwestern, NATO, Microsoft, and Canada’s CSE.

At 19, Ezz began his PhD in Computer Sci at the Univ of Waterloo, quickly gaining recognition through over 20 influential papers and open-source tools.

His experience includes leading advanced AI security ops projects for Orange CyberDefense, Forescout, RBC, and Huawei US.

He holds certifications such as GIAC Advisory Board, aCCISO, CISM, CRISC, GCIH, CEH, PMP and GCP-Cloud Architect, and served as an adjunct professor in cyber defense and warfare.

Creator: Middle Easterns & Africans in Cyber Security (MEACS)

Auto-condense telemetry volumes by 80% without data loss by grouping related records & deduping redundant values, with ML that auto-normalizes to a unified schema & auto-enriches data with relevant abstractive classifications, framework-aligned controls, contextual tags & environment labels. Use the efficiency with TKG ML to scale multi-level correlations & reveal all similarity & causality patterns & root causes. Use AI to automate troubleshooting, investigations, remediations & operations.

People:
    SpeakerBio:  Ezz Tahoun

Ezz Tahoun is an award-winning cybersecurity data scientist recognized globally for his innovations in applying AI to security operations.

He has keynoted, trained & presented at BlackHat US, Sector, MEA, Asia & EU, DEFCON, SANS Summits, all the top Bsides, Securityweek ICS Conference and GISEC among many others.

His groundbreaking work earned him a gold edison award and accolades from Yale, Princeton, Northwestern, NATO, Microsoft, and Canada’s CSE.

At 19, Ezz began his PhD in Computer Sci at the Univ of Waterloo, quickly gaining recognition through over 20 influential papers and open-source tools.

His experience includes leading advanced AI security ops projects for Orange CyberDefense, Forescout, RBC, and Huawei US.

He holds certifications such as GIAC Advisory Board, aCCISO, CISM, CRISC, GCIH, CEH, PMP and GCP-Cloud Architect, and served as an adjunct professor in cyber defense and warfare.

Creator: Blacks In Cyber Village

Explore how to leverage infrastructure OSINT data, such as passive DNS and domain registration records, to enhance your threat intelligence and protect your organization without exhausting budgets. In this demo, DomainTools Senior Security Advisor Malachi Walker will discuss how security teams can develop advanced techniques for continuous and explainable intelligence. Then, DomainTools Principal Product Manager Anthony Johnson will demonstrate how these concepts can be applied to build AI-driven product layers that draw domain intelligence from standard LLMs like ChatGPT, Claude, Gemini, and Copilot. Discover innovations that reduce triage time, improve early warnings, and automate campaign linkages, transforming your existing intelligence practices.

People:
    SpeakerBio:  Malachi Walker, DomainTools

Malachi Walker, DomainTools Senior Security Advisor, brings extensive experience in information security, spanning DNS, cyberspace crime and conflict, and cybersecurity governance and program design. They have previously worked in FTI Consulting s Cybersecurity practice and led product and brand protection efforts at WhiteHawk Inc. Malachi holds a Master s in Business Administration with a concentration in Cybersecurity Management from Virginia Polytechnic Institute and State University.

SpeakerBio:  Anthony Johnson, DomainTools

Anthony Johnson, DomainTools Principal Product Manager, has nearly 15 years of experience in cybersecurity. His background includes forensics, analytics, and product management, gained through work with the US Military, Casino Gaming, and other security-focused organizations.

Creator: Policy @ DEF CON

Between 2022 and 2024, the European High North saw three events at three attribution confidence levels. One segment of Svalbard’s twin fibre system failed—Norwegian police closed the case with no state-actor finding. AcidRain wiped tens of thousands of Viasat SurfBeam2 modems via a ground-segment management-plane compromise—attributed to GRU by US/UK/EU in May 2022. GNSS interference over eastern Finnmark persisted 2019–2024—assessed by Norwegian services as Russian EW largely spilling from Kola. Three attributions. Three regimes. No mechanism that treats them as one. UNCLOS cable protection varies by zone (Art 79 shelf; Art 113 high seas; Art 58 EEZ). The 1884 Cables Convention—still in force—governs outside territorial waters. Svalbard Treaty reach into the FPZ is contested. NATO’s cumulative-effects language has no classification framework. The 2023–24 Baltic sequence (Balticconnector; C-Lion1/BCS; Estlink-2/Eagle S) is the same play in the active theatre NATO CUI Cell, Baltic Sentry, EU CER/NIS2, JEF and NORDEFCO already cover — and still falls through. We propose an allies-first application of UNDRR Hazard Information Profiles to adversarial compound operations: indicators, authorities, a 2027 milestone. Classification stops no cable. A common record is the precondition for everything that will.

People:
    SpeakerBio:  Szymon Skalski, Jagiellonian University PL

Szymon is a PhD Candidate in the School of Social Sciences at Jagiellonian University and a Senior Expert and NASK National Research Institute in Poland. He is also affiliated with the Ostrom Workshop at Indiana University, where he leads the working group on Environmental Security and Technological Risk in Conflict. He has served on cybersecurity expert panels at the Polish Ministry of Digital Affairs and the Polish Bank Association. He is also a fellow of the European Law Institute, member of FIDMA and Virtual Routes European PhD Cybersecurity Accelerator Programme. For the past three years, he has participated in NATO’s Locked Shields exercises, representing Poland for two years and, in 2026, organizing the exercises on behalf of the CCDCOE in the legal domain.He is a researcher and scholarship holder of grants from the National Science Centre in Poland in the fields of insurance and digital security. He publishes in international academic journals in the fields of cybersecurity, insurance, tort law, and international law.

SpeakerBio:  Patricia Vargas Leon, Ostrom Workshop Indiana University US, Yale Law School

Patricia A. Vargas León is a Visiting Fellow at Yale Law School’s Information Society Project, a Cybersecurity Research Postdoctoral Fellow at Indiana University’s Ostrom Workshop, and a Visiting Scholar in Internet Governance at the Catholic University of Uruguay. Her research bridges law, policy, and technology — focusing on Internet governance, cybersecurity, international law, and the law of the sea. She is particularly interested in how governments control Internet infrastructure, network neutrality, and regulatory parallels between the law of the sea and cyberspace. Her doctoral dissertation examines Internet blackouts across political regimes and how states restrict digital connectivity. Before academia, Patricia practiced law in the private sector for nearly a decade and brings over two decades of international law experience to her work, including a consultancy with the UN Division for Ocean Affairs and the Law of the Sea (DOALOS). She also served as a Google Policy Fellow during her doctoral studies. Patricia holds a Ph.D. and M.S. in Information Science and Technology from Syracuse University and a law degree from the Pontifical Catholic University of Peru. Her dual training positions her to address complex questions at the intersection of technology, governance, and international law.

SpeakerBio:  Jorge Acevedo Canabal, Ostrom Workshop Indiana University US

Jorge Acevedo Canabal, MD (University of Puerto Rico School of Medicine, Magna Cum Laude), is a physician and Visiting Scholar at the Ostrom Workshop, Indiana University, working on research that sits at the intersection of healthcare, public health, and cybersecurity, applying epidemiological and disaster medicine methods to map patient harm attributable to healthcare cyberattacks and technological hazards. He previously served as Chief Medical Officer of the Puerto Rico Science, Technology and Research Trust, and currently serves as advisor to the Biohacking Village and Raíces Cyber Org.

Creator: AppSec Village

Container escapes sound scary and complicated. In practice, most of them come down to a handful of misconfigurations that show up in production environments all the time. Overprivileged containers, exposed Docker sockets, weak seccomp profiles, and poorly scoped capabilities. This talk demystifies container escapes for AppSec practitioners. We will walk through three real escape techniques live inside Docker environments running on Google Cloud, explain exactly why each one works, and then show the configuration changes and detection controls that prevent them. No black magic. Just misconfigurations you can find and fix right now, in environments that look a lot like yours.

People:
    SpeakerBio:  Advait Patel

Advait Patel is a Senior Site Reliability Engineer at Broadcom and the creator of DockSec, an open-source, AI-powered Docker security analyzer.

Advait is a Docker Captain, Google Developer Expert and regular speaker at major security and developer conferences, including QConAI Boston, SANS CloudSecNext, OWASP GlobalAppSec US, OWASP SnowFROC, PlatformCon, Linux Security Summit North America, O’Reilly Media AI CodeCon, Redgate Cloud Summit, CornCon, ISACA, Silicon Valley Cybersecurity Conference.

His work sits at the intersection of AI, security, and developer workflows, with a particular focus on how emerging technologies like LLMs are reshaping modern development, CI/CD pipelines, and Git-based collaboration. As part of the Git and DevOps community, he advocates for secure-by-default design, developer empowerment, and open-source tooling that puts security in the hands of those who build the future.

Creator: Red Team Village

Demos (Desktop Only at the moment): https://doctoreww.github.io/EvilFontTool/

Tool: https://github.com/DoctorEww/EvilFontTool

General: Workshop goes into evil fonts and how to use them in a red team to get free shells, also hints at a tool to make it easier (saving most of the tool stuff for the tactic)

Tactic will go into using the tool to create malicious word docs for social engineering / popping a shell.

Description: For decades, cybersecurity has taught us to validate what we click, what we download, and what we run. Analysts learn to inspect URLs carefully, read every line, and challenge every prompt or popup that doesn’t look quite right. But in all this vigilance, we’ve ignored a foundational assumption. We trust our eyes. We trust that what we see is what the system sees. This talk demonstrates why that assumption puts us at risk.

In the modern computing stack, very little sits closer to human perception than fonts. They are in browsers, PDFs, document editors, terminals, IDEs, chat apps, and nearly every interface in between. Yet despite their ubiquity, fonts are almost never treated as part of the security boundary. They’re handled as passive design elements, harmless vessels used solely to shape text. Users trust them implicitly. This session introduces a new class of “evil font” attacks. With evil font attacks, red teamers can create situations where a user believes they are copying, clicking, or executing one thing… while the machine processes something entirely different.

Evil font attacks don’t depend on JavaScript, macros, or browser exploits. Evil fonts work because there is no reason that a specific letter needs to look the way it does. In other words, the letter “A” only looks like an “A” because that’s how my font drew it. With clever font manipulation a red teamer can make the letter “A” look like any other letter. For example, while a user might see the letter “B” the computer would still see the letter “A” represented by the underlying ASCII hex value. By manipulating fonts, files no longer have a single meaning… They mean one thing to a human reader, and another thing to a computer.

Once you realize this trust boundary is broken, there are so many ways to abuse this.

During the session, we walk through three attack scenarios usable by red teams; each scenario highlights a different security boundary where fonts exert control.

Clipboard deception in modern browsers Clipboard attacks typically require JavaScript, event hijacking, or social engineering. With evil fonts you can poison a clipboard on webpages without any JavaScript. With evil fonts an attacker can make on-screen text appear safe (e.g., “echo hello world” or “https://google.com”) while the underlying copied text is entirely different. Users believe they are copying the text they verified with their eyes, but instead the clipboard receives text the attacker controls.

Document poisoning: Like browsers, documents like docx and pdf files can by poisoned by evil fonts. In this portion of the talk, we explore how to poison word and pdf files with malicious links and commands. These documents can be emailed for deception based social engineering. More insidious, evil fonts can be used to poison existing documents within a target such as how to’s and infrastructure guides. By poisoning documents, red teamers can turn simple shared drive access into credentials and shells.

AI‑era submission manipulation: The rise of AI‑powered document processing introduces a new frontier for deception. Modern systems summarize text, extract entities, detect sentiment, identify anomalies, and evaluate authenticity. Evil fonts can exploit this by making the text look safe to the computer… but when read by a user, it displays malicious content. This manipulation can skew automated summaries, risk scoring, keyword extraction, plagiarism detection, and content filtering.

Please let me know if you have any questions! Thanks, Andrew

People:
    SpeakerBio:  Andrew Griess

Andrew Griess is a Red Teamer at Centene who gets paid to break things and calls it a career. With 3 years of professional red team experience, he’s made it their mission to think like an attacker while not succumbing to the dark side. When not poking holes in things for money, Andrew is deep in security research — chasing the next interesting bug, developing new techniques or going down rabbit holes that started as “just a quick look.”

Creator: Biohacking Village

Ransomware Kills Patients. We Can’t Prove It. Here’s How We Fix That.

Ransomware attacks on hospitals kill people. That’s not a hypothesis — it’s in the data. A peer-reviewed analysis of Medicare claims (American Economic Journal: Economic Policy, 2026) puts in-hospital mortality at 34 to 38 percent higher during attacks. The dead are disproportionately elderly, critically ill, and patients of color. Dameff et al. (2023) documented emergency department spillover. Neprash, Dameff, and Tully (2024) traced the same pattern through the Change Healthcare attack.

The mechanism isn’t exotic. Encrypted EHRs mean clinicians are flying blind — no medication history, no imaging, no labs. Networked infusion pumps, ventilators, and monitors drop to manual. Ambulances get diverted, stacking patients at facilities that weren’t hit. In rural areas, transfer times go from nine minutes to thirty-three. Here’s the deeper problem: these deaths are architecturally invisible.

No ICD code exists for “died because the hospital’s network was encrypted.” No death certificate asks whether the hospital was under cyberattack. No public health surveillance system captures excess mortality from clinical failure caused by ransomware. Mandatory reporting requirements attach to data breach — not patient harm. Voluntary harm-reporting channels exist, but they’re anonymous and sporadic. The visible signal is a fraction of what’s actually happening.

This isn’t a technical gap. It’s a design failure in the detection infrastructure itself.

This talk makes the case for cyber-harm epidemiology. Using a forthcoming law review article, we show that existing systems — ICD external-cause coding, NCHS disaster-death certification, SNOMED CT alignment, the Sendai Framework’s Hazard Information Profiles (2025) and Global Disaster-Related Statistics Framework (2026) — can be adapted right now to make cyber-attributable patient deaths visible at population scale. No new treaties required.

Better detection produces better attribution. Better attribution makes state obligations under the right to life and the protection of medical units enforceable — not aspirational.

The deaths are real. The tools to count them exist. We just haven’t connected them yet. That’s what this talk is about.

People:
    SpeakerBio:  Jorge Acevedo Canabal, Ostrom Workshop Indiana University US

Jorge Acevedo Canabal, MD (University of Puerto Rico School of Medicine, Magna Cum Laude), is a physician and Visiting Scholar at the Ostrom Workshop, Indiana University, working on research that sits at the intersection of healthcare, public health, and cybersecurity, applying epidemiological and disaster medicine methods to map patient harm attributable to healthcare cyberattacks and technological hazards. He previously served as Chief Medical Officer of the Puerto Rico Science, Technology and Research Trust, and currently serves as advisor to the Biohacking Village and Raíces Cyber Org.

SpeakerBio:  Scott Shackelford

Scott J. Shackelford is Associate Vice President and Vice Chancellor for Research at Indiana University Bloomington and Provost Professor of Business Law & Ethics at the IU Kelley School of Business. He serves as Executive Director of both the Ostrom Workshop and the Center for Applied Cybersecurity Research, and directs the Ostrom Workshop Program on Cybersecurity & Internet Governance. He is also an Affiliated Scholar at Harvard Kennedy School’s Belfer Center and Stanford’s Center for Internet and Society. Scott has authored over 100 articles, book chapters, and essays, with research featured in Politico, NPR, CNN, Forbes, Time, and the Washington Post. His books include The Internet of Things: What Everyone Needs to Know (2020) and Managing Cyber Attacks in International Law, Business, and Relations (2014). His honors include a Harvard Research Fellowship, a Stanford Hoover National Fellowship, the 2015 Elinor Ostrom Award, and the 2022 Poets & Quants Best 40-Under-40 MBA Professors Award.

SpeakerBio:  Szymon Skalski, Jagiellonian University PL

Szymon is a PhD Candidate in the School of Social Sciences at Jagiellonian University and a Senior Expert and NASK National Research Institute in Poland. He is also affiliated with the Ostrom Workshop at Indiana University, where he leads the working group on Environmental Security and Technological Risk in Conflict. He has served on cybersecurity expert panels at the Polish Ministry of Digital Affairs and the Polish Bank Association. He is also a fellow of the European Law Institute, member of FIDMA and Virtual Routes European PhD Cybersecurity Accelerator Programme. For the past three years, he has participated in NATO’s Locked Shields exercises, representing Poland for two years and, in 2026, organizing the exercises on behalf of the CCDCOE in the legal domain.He is a researcher and scholarship holder of grants from the National Science Centre in Poland in the fields of insurance and digital security. He publishes in international academic journals in the fields of cybersecurity, insurance, tort law, and international law.

Creator: Crypto & Privacy Village

Confidential computing systems involve computation over private inputs held by mutually distrusting parties while producing public, cryptographically verifiable evidence of what was computed. Trusted execution environments (TEEs) are a practical building block for these systems, enabling code to run inside hardware-encrypted enclaves that emit attestations binding a measurement of the loaded code to genuine hardware. We present Cove, a framework that composes attested computations into multi-stage, multi-party workflows structured as directed acyclic graphs. Each node runs in its own enclave, decrypts private inputs only under attestation-gated key release, and emits a certificate that downstream nodes can require as a cryptographic precondition before consuming upstream artifacts; anyone holding the workflow bytes and the TEE vendor’s attestation roots can verify the whole chain non-interactively. We show that a small set of reusable primitives – confidential artifact provisioning, encrypted dynamic outputs, attested ephemeral-key channels (RA-TLS), and certificate-gated preconditions – compose into systems that lift attestation from verified code identity to verified runtime properties, and use this to express practical applications in secure AI systems including confidential AI inference, attested AI benchmarks, and training-code verification. We give an open-source reference implementation on Intel TDX via Phala Cloud’s dstack and demonstrate end-to-end feasibility with an attested benchmark workflow.

People:
    SpeakerBio:  Stephanie

Stephanie is a research engineer working on AI security and safety. Her research interests include adversarial robustness, agent security and verifiable AI deployments. Previously, she was at Meta, where she worked on AI safety and security evaluations (CyberSecEval) and guardrails (PromptGuard, LlamaFirewall), and prior to that she worked on application security for web and mobile.

SpeakerBio:  Robin

Robin is a Member of Technical Staff at Inferact (maintainer of vLLM), working towards the best inference stack of the future. Previously he worked as a software engineer at Near Protocol, Meta, Google, and Twitter. His interests include application security, model red-teaming, distributed systems, and GPU performance optimization.

SpeakerBio:  Erika Lee

Erika Lee is a 3rd-year Math–Computer Science undergraduate student at UC San Diego and a MATS 9.1 AI Security Fellow. Her research interests are in verifiable and privacy-preserving AI, and building secure and trustworthy AI systems. She previously interned with the US Department of Defense’s AI/ML portfolio and the U.S. Army, working on deploying AI/ML systems in classified environments.

Bing-Jyue Chen is a 3rd-year PhD candidate in Computer Science at UIUC. His current research focuses on zero knowledge machine learning (zkml). He is also interested in verifiable AI, privacy-preserving machine learning, and advanced cryptography.

Daniel Kang is an assistant professor at UIUC in the computer science department. His research focuses on making analytics with machine learning easy for scientists and analysts to use, and has been supported by Google, the Open Philanthropy project, Emergent Ventures, and others. He has consulted at OpenAI, Google, Microsoft, hedge funds, and other companies to support their ML deployments, and advised a number of startups.

Creator: Quantum Village

Quantum computers will inevitably move to multi-tenant systems for efficiency and throughput, with many users sharing a single QPU. This talk will discuss the realities of threats facing quantum computing ‘as a service’ and what is particularly specific to quantum computing. We’ll get you started on understanding the quantum computing tech stack, and work through a few examples of possible attacks and mitigation techniques, wrapping up our newly developed security paradigm of covert quantum computing. All of the examples presented in this talk can be run by anyone with an internet connection and an email address. This talk is for anyone curious about attacks (and mitigation) techniques in quantum computing, and similarly how to build and test new ideas. A quantum background is not required!

People:
    SpeakerBio:  Evan Anderson, UMD

Evan is a postdoctoral researcher at the University of Maryland focusing on photonic (and atomic) quantum computing architectures.

Creator: .EDU Community
Giveaway: Two raffles for one copy of:
See Yourself in Cyber: Security Careers Beyond Hacking

People:
    SpeakerBio:  Shavvon “TheSiren” Cintron

Shavvon Cintron spent the last decade climbing the IT ladder, now working in IT and cybersecurity within the healthcare space. She’s got an MS in Cybersecurity from Pace, a BS in Criminal Justice from CUNY John Jay, and a 2025 SANS Institute Difference Makers Rising Star nod, so somebody thinks she’s doing something right. When she’s not at her day job, she’s Director of Women’s Education for the Women’s Society of Cyberjutsu, building the programs she wishes existed when she was trying to break in. She goes by “Siren of the Cyber Sea.” Get too close and you might not want to leave.

Creator: Malware Village
Abstract:
What happens when a leaked malware builder suddenly opens the door for anyone — from low-tier criminals to advanced operators — to deploy a backdoor armed with a kernel-capable rootkit? What happens when that rootkit still carries valid signatures, and Windows 11 willingly loads it?

In this talk, Check Point Research unveils a comprehensive reverse engineering of ValleyRAT, a modular backdoor family also known as Winos/Winos4.0. By dissecting the publicly leaked builder and its development artifacts, we reconstruct the malware’s architecture from the ground up, uncovering a highly organized plugin ecosystem and coding style that reveals a small, specialized team with deep knowledge of the Windows kernel.

The standout finding: ValleyRAT’s Driver Plugin, containing a stealthy and surprisingly robust kernel-mode rootkit. Despite being abused by multiple operators, several variants remain validly signed, bypassing Microsoft’s driver protections and loading successfully on the latest Windows builds. The rootkit’s functionality — including coercive AV/EDR driver deletion, silent installation methods, and APC-based shellcode injection — demonstrates a level of sophistication rarely seen in malware distributed through public builders.

Our telemetry and detection work further expose a dramatic rise in ValleyRAT’s presence in the wild. Roughly 85% of detected plugin samples surfaced within the past six months, directly following the builder leak. This surge marks a turning point: ValleyRAT is no longer a tool reliably associated with specific Chinese-speaking threat actors, but rather an accessible platform fueling opportunistic and un-attributable campaigns.

This research pulls back the curtain on ValleyRAT’s internals, its kernel rootkit mechanics, and the implications of a powerful multi-module malware ecosystem becoming available to the masses. When sophisticated tooling escapes into the wild, who gets to wield it — and who pays the price?

People:
    SpeakerBio:  Jiří Vinopal, Threat Researcher at Check Point Research

Jiří Vinopal is a security researcher, malware researcher, and reverse engineer at Check Point Research, focused on advanced cyber threats, kernel internals, and the hidden mechanics of undocumented system components. His work spans uncovering novel attack primitives, reconstructing proprietary protocols from binary analysis alone, and deep-diving into both sophisticated malware families and trusted platform components. When he’s not buried in disassembly, he actively shares his knowledge and passion for reverse engineering across his X account, YouTube channel, and blog — delivering tips, tricks, and technical insights to fellow enthusiasts and the broader security community.

Creator: Red Team Village

Traditional web crawlers are great at finding links but terrible at understanding websites. The interesting challenge is not replacing crawlers with AI. It’s using the minimum amount of AI required to dramatically improve coverage.

Every automated tool I’ve tried hit a brick wall the moment it encounters MFA, multi-step workflows, onboarding wizards, or application states where a wrong click deletes data. As a result, large portions of authenticated attack surface require manually using the application.

I built a crawler that combines deterministic tooling such as Katana with a minimal AI agent. The AI is only used where traditional crawlers consistently fail: authentication, workflow navigation, and generation of safety controls. Once authenticated session state is established, the system hands control back to fast deterministic tooling.

I’ll cover the basics of how it’s built and we’ll try it together on a variety of applications.

People:
    SpeakerBio:  Daniel Goldberg

Daniel is a security researcher who spent the last 20 years crawling his way up the stack from hacking operating systems to attacking client-side applications to attacking browsers, network protocols, and today web applications. After dabbling with everything in security, he’s realized what he really enjoys is how to take something that works and make it really good

Creator: Scambait Village

Join popular scambait content creators RinoaPoison and VanHelen for an open, behind-the-scenes conversation about live scambaiting, character work, community building, and the realities of turning the tables on scammers on Twitch and YouTube. Expect stories from the calls, tips for aspiring creators, and insights into keeping the work sustainable and entertaining while still disrupting scam operations.

Links:
    scambaitvillage.org/speakers – https://scambaitvillage.org/speakers

People:
    SpeakerBio:  RinoaPoison, Twitch Streamer / Content Creator
No BIO available
SpeakerBio:  VanHelen, Twitch Streamer / Content Creator
No BIO available

Creator: Crypto & Privacy Village

Developers keep getting crypto wrong in the same ways. Not the math, but the code around it. A missing check here, implicit trust there, and suddenly a signature means nothing. This talk is data-driven and hands-on. Starting from OWASP A04:2025, we examine real CVEs from GitHub Security Advisories as of January 2026 to show which crypto failure patterns actually dominate in the wild, then go deep on the top two: signature verification bypasses (including invalid curve attacks) and algorithm confusion bugs (JWT). Real libraries, live exploits. Demos and CTF challenges are woven throughout, involves audience interaction and participation. No crypto background required.

People:
    SpeakerBio:  Diptendu Kar

Diptendu Kar is a security researcher focused on supply chain and dependency risk. He works on triaging open-source vulnerabilities, writing detection rules, and exploring how AI can automate tedious parts of security research. He also teaches Software Security Practices at Northeastern University part time and holds a Master’s in Cybersecurity. Before security, he worked as a Java developer at TCS. He is especially interested in patch diffing, vulnerable function detection, and the use of LLMs in AppSec. He believes cryptography is magic and AI is a noisy intern – helpful but always needs supervision.

Creator: Cryptocurrency Village

Closing the Cryptocurrency areas at DEFCON, we introduce everyone at work and their achievements during the event. We review our four major activities including the hackathon, contests, workshops, and capture the flag competition. A summary of teams, winners, and statistics indicates the degree of participation in this event. We describe what prizes were awarded and which contestants won big. A list of workshop topics follows, with instructors giving their impressions of how modern finance technology benefits from interactive learning. Finally, we give a sneak preview of what’s to come in the Cryptocurrency areas (Village, Contest, Vendor, Party, Training) at DEFCON in Bahrain and how we intend to fulfill the mandate of fostering cryptocurrency exploration and hacking there.

People:
    SpeakerBio:  Chelsea Button, Cryptocurrency Education Initiative

Chelsea is a lawyer specializing in consumer finance, data and technology. She advises clients on updates in the law and defends them in litigation. She is a cryptocurrency advocate, with multiple professional publications.

SpeakerBio:  Param “P7R7M”

Param is an Electrical Engineering Student from Georgia Tech with a strong passion for and interest in crypto. Although he primarily got interested in cryptography and hardware security through a class at Georgia Tech, he is also working at a software company on crypto adoption and ease of use. With a unique blend of HW and SW skills, Param is truly enthusiastic about all aspects of crypto.

SpeakerBio:  Arjun “Peper” Suresh, Postgraduate Student, University of Wollongong in Dubai

Arjun Suresh is pursuing postgraduate studies in Cybersecurity at the University of Wollongong in Dubai, specializing in blockchain security, penetration testing, and applied cryptography. His interest in crypto security was shaped by analyzing major exchange breaches, including the Mt. Gox and Ronin Network hacks, where he studied the gap between attacker tradecraft and real-world defenses.

Creator: Cryptocurrency Village

Join your fellow hackers managing the Cryptocurrency areas at DEF CON, and get a sneak peak of what each workshop teaches as well as an overview of the showcases and programs happening in our DEF CON Village and Contest areas. We will report on cryptocurrency trends and perspectives from distinguished positions in industry, academy, and government. We will announce the teams competing in the Cryptocurrency Challenge, and give an overview of what prizes are available to winning contestants of the hackathon, CTF, and contests. Meet the organizers of years of cryptocurrency content at Defcon and bring your questions to the Creator Stage during the Cryptocurrency Opening Keynote!

People:
    SpeakerBio:  Michael “MsvB” Schloh, Chairman, Monero Devices

The Cryptocurrency Advocate is a group working to prepare society for the likely adoption of modern cryptocurrency in legacy financial systems. We host a number of events, including the Cryptocurrency Village and Cryptocurrency Challenge at DEFCON hacker conventions and other cybersecurity events around the world.

SpeakerBio:  Param “P7R7M”

Param is an Electrical Engineering Student from Georgia Tech with a strong passion for and interest in crypto. Although he primarily got interested in cryptography and hardware security through a class at Georgia Tech, he is also working at a software company on crypto adoption and ease of use. With a unique blend of HW and SW skills, Param is truly enthusiastic about all aspects of crypto.

SpeakerBio:  Arjun “Peper” Suresh, Postgraduate Student, University of Wollongong in Dubai

Arjun Suresh is pursuing postgraduate studies in Cybersecurity at the University of Wollongong in Dubai, specializing in blockchain security, penetration testing, and applied cryptography. His interest in crypto security was shaped by analyzing major exchange breaches, including the Mt. Gox and Ronin Network hacks, where he studied the gap between attacker tradecraft and real-world defenses.

Creator: Cryptocurrency Village

In this hour, we’ll guide you from setup to exploitation of a Cryptohack electronic badge. From installing the ESP-IDF toolchain, understanding application logic, compiling and disassembling firmware, we get familiar with the basics before considering the advanced aspects of financial technology in hardware. To explore the typical interfaces of a hardware wallet, we experiment with live hacks on a testnet wallet supporting Solana, Ethereum, Bitcoin, and Monero. By the end of the session, you’ll know how to probe the boundaries of hardware security and leave with insights you can apply to real-world offensive security challenges.

People:
    SpeakerBio:  Do Truong Giang “FSNaix”, BlossomsAI / Bloomify (AI & Security)

I’m Giang (FSNaix), an AI and security researcher from Vietnam and a Bloomify cofounder. I help build open-source Vietnamese language models with BlossomsAI, I built Petal (an AI tool that reverse-engineers binaries faster than doing it by hand), and at DEFCON Singapore 1 I managed to get DOOM running on a $20 conference badge.

SpeakerBio:  Sadiq “Sdqmd”, CodeBit

sdqmd (Sadiq) is a hardware hacker and co-founder of CodeBit, a technology company based in Brunei that’s building a hardware engineering academy to teach embedded systems and security from the ground up. He is focused on developing a pedagogy that blends theory with hands-on practice, so aspiring hackers — especially those just starting out — don’t just learn how something works; they build it, break it, and understand it for themselves. His own research focuses on hardware attacks against embedded devices — using fault injection and power analysis to defeat the security of everything from crypto hardware wallets to point-of-sale infrastructure. Above all, he’s driven by one goal: building hardware and security competency in Brunei, where hands-on learning is still hard to come by.

SpeakerBio:  Arjun “Peper” Suresh, Postgraduate Student, University of Wollongong in Dubai

Arjun Suresh is pursuing postgraduate studies in Cybersecurity at the University of Wollongong in Dubai, specializing in blockchain security, penetration testing, and applied cryptography. His interest in crypto security was shaped by analyzing major exchange breaches, including the Mt. Gox and Ronin Network hacks, where he studied the gap between attacker tradecraft and real-world defenses.

Creator: OSINT For Good Community

Participating in today’s Global OSINT Search Party CTF? This session, offered by our most senior Coach, the Legendary KennyJ, will do a quick overview of what to expect, how to get organized, and how to make sure you’re following the rules.

People:
    SpeakerBio:  Kenny J, Trace Labs

Senior Infrastructure Engineer by day. Osint for Good by night. His is the only Trace Labs coach to have coached 20+ CTFs, earning him the singular rank of Legendary Coach.

Creator: Noob Community
Session Description: Technical skills might open doors in IT & Cybersecurity, but leadership, communication, and mentorship sustain growth and impact. As the field becomes more complex and critical, we must rethink how we build and lead teams. This talk highlights the soft skills that turn strong technicians into effective collaborators, trusted advisors, and emerging leaders: from coaching junior analysts to translating between the SOC and the C-suite.
Takeaways: Attendees will learn practical ways to foster teambuilding, conflict resolution, psychological safety, build mentorship pipelines, and communicate with clarity under pressure.
Target audiences: Everyone

People:
    SpeakerBio:  Amanda Kollmorgan
No BIO available

Creator: Blacks In Cyber Village

As artificial intelligence shifts from supporting analytics to exercising real-time operational control in critical infrastructure energy grids, aviation systems, healthcare networks, transportation, and national logistics it ceases to be a tool and becomes part of the infrastructure itself. AI now optimizes load balancing, automates routing, manages predictive maintenance, and drives high-stakes decisions that directly impact public safety, economic stability, and national security. This session introduces the CTRL framework (Control, Resilience, Trust, Leadership) a strategic blueprint for securing autonomous systems before they evolve into systemic risk. Drawing from hands-on experience in federal cybersecurity, aviation security, AI governance, and academic cybersecurity instruction, Nykolas Muldrow explores how AI expands the attack surface through threats like data poisoning, model drift, prompt injection, and compromised AI supply chains. Attendees will gain: A clear view of how embedded AI transforms infrastructure vulnerabilities A practical threat model tailored to autonomous operational systems Governance strategies aligned with NIST AI Risk Management Framework and ISO-based controls The CTRL leadership model for embedding responsible AI oversight into organizational strategy An immediate-use risk-evaluation method for their own products, infrastructure, or agencies In an era where regulators intensify scrutiny, investors demand deeper risk diligence, and operators race to deploy AI without proportional safeguards, the next decade of progress will belong to those who govern autonomous systems responsibly not just those who deploy them fastest. Join to equip yourself with the tools to lead before AI makes decisions we cannot easily reverse.

People:
    SpeakerBio:  Nykolas Muldrow, CEO of CI Solutions Global Inc.

Nykolas Muldrow is a cybersecurity executive, CEO of CI Solutions Global Inc., and 2025 GovTechCon Mission Trailblazer Award recipient for their pioneering work in AI, cybersecurity, and secure cloud technologies for critical infrastructure. With 15+ years securing defense, aviation, and federal systems including aviation GRC at American Airlines they specialize in threat modeling autonomous AI in high-stakes environments like energy grids, transportation, and healthcare networks. Nyk teaches cybersecurity, aligns solutions to NIST AI RMF and ISO controls, and equips leaders to govern AI responsibly before it becomes systemic risk.

Creator: Blacks In Cyber Village

Join B.I.C. Village for Cyber Gamechangers: Women Who Lead, Secure, and Inspire, a welcoming session focused on leadership, career growth, and the impact of women shaping cybersecurity. Come ready to learn, ask questions, and connect with the community.

People:
    SpeakerBio:  Nikkia Henderson, Senior Advisor, Cybersecurity Infrastructure Security Agency (CISA) / President, Women in Cybersecurity Mid Atlantic Affiliate

Nikkia Henderson is a Senior Advisor at the Cybersecurity Infrastructure Security Agency (CISA), where they lead the Cyber Supply Chain Risk Management (C-SCRM) Strategy and Governance Program. They also serve as President of the Women in Cybersecurity Mid Atlantic Affiliate, helping professionals define their vision and career paths in federal cybersecurity. With over 14 years in federal government, Nikkia holds a Master’s degree in Cyber Policy.

SpeakerBio:  Arielle Baine, Chief of Cybersecurity at Cybersecurity and Infrastructure Security Agency (CISA)

Arielle Baine is the Chief of Cybersecurity for Region 3 within the Department of Homeland Security’s Cybersecurity and Infrastructure Security Agency (CISA). They lead the Cybersecurity Advisor Program, enhancing the security and resilience of critical infrastructure. With over 9 years of experience in cybersecurity across federal civilian and DoD communities, Arielle holds multiple certifications including CISSP, CCSP, GCWN, CEH, and Security+, and a Master of Science degree in Cybersecurity.

SpeakerBio:  Jess Hoffman, Deputy CISO, City of Philadelphia / Professor, Harrisburg University and The Pennsylvania State University

Jess Hoffman is a Certified Information System Security Professional (CISSP) with nearly 20 years of IT and cybersecurity experience in government and private sectors, focusing on Audit and Compliance. They currently serve as Deputy CISO for the City of Philadelphia and are a Professor at Harrisburg University and The Pennsylvania State University. Jess is a national speaker and advocates for mentorship and DEI within cybersecurity.

Creator: Blue Team Village

This talk is a fast-paced, vendor-neutral primer for defenders who want to turn raw telemetry into actionable intelligence. Threat intelligence is not just a feed — it is a discipline built through practice, structure, and action. This session introduces the CTI lifecycle, intelligence requirements, MITRE ATT&CK mapping, and the role of baselining in helping analysts separate normal activity from suspicious behavior.

Attendees will also see how CTI becomes operational through practical blue-team workflows such as YARA and Sigma rule development, threat hunting, incident response, and purple-team validation. We will close by exploring how grounded AI workflows can accelerate enrichment, ATT&CK mapping, detection drafting, and analyst reporting while keeping humans responsible for validation, context, and final decisions.

People:
    SpeakerBio:  Carlo Anez Mazurco

Carlo Anez Mazurco is a cybersecurity instructor, consultant, and community leader with more than 15 years of experience across security operations, threat intelligence, incident response, threat hunting, and detection engineering. He is the Founder of IgniteCyber Academy, a DEF CON Training instructor, and an active contributor to Blue Team Village, where he supports Project Obsidian and develops hands-on blue team content for the cybersecurity community.

Carlo specializes in helping defenders translate attacker tradecraft into practical detection and response techniques while responsibly integrating artificial intelligence into modern security operations. His work focuses on creating realistic labs, CTF challenges, and immersive training environments that prepare students for real-world investigations using enterprise telemetry, cloud technologies, and AI-assisted workflows.

He has delivered training and presentations for conferences, universities, government organizations, and commercial teams, with a passion for mentoring the next generation of cybersecurity professionals. His goal is to make complex security concepts approachable through practical demonstrations, collaborative learning, and hands-on exercises that participants can immediately apply in their own environments.

Creator: Recon Village

What if your recon tool could understand “find leaked credentials for this company on the dark web, cross-reference with their exposed infrastructure, and show me the attack path” — and then actually do it, autonomously, with zero manual commands? CyberKB is a 214,000-line open platform that puts an AI copilot (Keke) in command of 131 offensive tools spanning reconnaissance, dark web OSINT, Shodan intelligence, breach databases, and a full Kali Linux shell — all orchestrated through natural language conversation. The Dark Web Intelligence Pipeline (DarkMonitor): CyberKB’s DarkMonitor module queries .onion search engines concurrently through Tor, uses LLM-powered query refinement to generate optimal dark web search terms, applies AI relevance filtering to surface the most critical results.

The AI Copilot (Keke): Keke isn’t a wrapper around ChatGPT. It’s a function-calling agent with direct execution access to: a privileged Kali container (nmap, curl, nikto, sqlmap, gobuster, hydra — the full toolkit), 16 dark web search engines via Tor, breach and paste database aggregators, a RAG-powered knowledge base with semantic search over ingested recon data, Shodan lookups and CVE correlation, MITRE ATT&CK technique mapping, attack graph generation and path prioritization, and engagement management (targets, credentials, findings, reports).

Scale Proof — 2,250-Domain Assessment: We demonstrate CyberKB against a real-world external attack surface assessment: 2,250 domains belonging to a single organization, producing 11,968 unique IPs, 8,494 hostnames, 2,444 CVEs, 80,238 open ports, and a knowledge graph of 6,390 nodes with 51,990 infrastructure relationship edges. The entire dataset was parsed, correlated, and ingested into the AI-queryable knowledge base in 25 seconds. Keke can now answer questions like “which x domains share infrastructure with known-vulnerable IPs” by searching the graph, not by re-scanning.

What This Means for Defenders: Every autonomous recon step Keke performs leaves detectable artifacts. We discuss what blue teams should monitor: DNS burst patterns from multi-engine enumeration, Tor exit node correlation with target infrastructure, behavioral signatures of AI-driven sequential probing, and how to distinguish human recon from autonomous agent recon. The same platform that empowers red teams reveals the detection surface that defenders can leverage.

Key Takeaways: 1. AI copilots with direct tool execution compress hours of reconnaissance into minutes of conversation — fundamentally changing the operator’s role from command executor to strategic decision-maker. 2. Dark web OSINT can be systematically automated with LLM-driven search refinement and relevance filtering, making it accessible beyond specialized analysts. 3. Infrastructure-scale attack surface mapping (2,250+ domains) becomes practical when AI handles correlation instead of humans. 4. Autonomous recon creates detectable patterns that blue teams should be actively hunting for. Tool Stack: Python, Flask, ChromaDB (RAG), Docker (Kali + Tor + Browser Agent), OpenAI-compatible LLM API, SQLite, Playwright, BeautifulSoup. 105 Python modules. Fully self-hosted — no cloud dependencies for core functionality.

People:
    SpeakerBio:  Suriya Prasath S, zoho, red teamer, manager

A Red Teamer and Security Manager at Zoho CRM–Red Team, with 6+ years of experience driving adversarial simulations and real-world attack emulation at scale. He focuses on uncovering critical security gaps by thinking like an attacker—blending deep technical expertise with practical red team operations to challenge assumptions and strengthen defences.

SpeakerBio:  Chandru J, Zoho, Product Security Manager

Product Security Manager with over 12 years of experience in driving cybersecurity initiatives, enhancing organizational security posture, and ensuring compliance with international standards such as ISO 27001, SOC 2, and GDPR. Adept at leading and mentoring teams, managing enterprise-wide security programs, and developing in-house security tools to address vulnerabilities effectively. Proven expertise in vulnerability management, incident response, security governance, and implementing secure development lifecycle (SDLC) practices. Recognized for fostering a security-first culture and collaborating with cross-functional teams to mitigate risks, protect assets, and improve processes in rapidly evolving environments.

SpeakerBio:  Muthu Kumar, Security Engineer

I am a Security Engineer with 10 years of experience specializing in web application security and security tool development. I have extensive experience identifying security vulnerabilities, improving secure development practices, and building tools that help organizations detect and prevent security risks at scale.

My expertise includes application security testing, secure code reviews, threat modeling, vulnerability assessment, and developing security automation solutions that reduce manual effort and improve detection accuracy. I have worked on enhancing security tools by reducing false positives and helping engineering teams address vulnerabilities more efficiently.

I am passionate about building secure systems and solving complex security challenges through automation, innovation, and practical security engineering approaches. My focus is on strengthening application security while enabling development teams to build and ship secure products faster.

Creator: Biohacking Village

Doctors are burned out and turning to LLMs — uploading discharge letters, lab reports and papers into ChatGPT, Claude and clinical AI tools. Every PDF is an unvalidated input to a system that cannot tell real facts from injected lies. We introduce Divergent Prompt Injection: unlike classic attacks (“”ignore previous instructions””), divergent payloads are clinically plausible false statements — a wrong drug, a fabricated contraindication, a phantom diagnosis — embedded in medical content. They create no semantic conflict, trigger no detection system, and sit indistinguishable from real medicine until an LLM summarizes them into a treatment plan.

We release DarkSyringe, an open-source tool automating the full attack chain: PDF ingestion, de-identification, LLM-driven payload generation, injection and multi-model scoring. In testing, a single poisoned discharge letter caused multiple commercial LLMs to recommend wrong drugs, fabricate contraindications and invent diagnoses — errors that would directly harm patients.

This talk brings a physician’s perspective: understanding why even a low-complexity attack becomes critical when it lands in a clinical environment built on time pressure, trust, and information overload.

People:
    SpeakerBio:  Francesco Costa

Hand surgeon trainee and cybersecurity researcher focused on LLM security. Creator of DarkSyringe, combining real clinical experience with offensive research to expose the limits of current defenses.

Creator: Noob Community

This session is an open on building a career in offensive security, from unconventional beginnings to breaking AI systems at a major tech company.

The speaker will share a candid look at his journey from U.S. Army Special Forces to national-level cyber operations and now AI penetration testing highlighting what actually mattered, what didn’t, and how skills from completely different fields translate into high-end offensive security roles.

Topics will range from breaking into cybersecurity with no experience, navigating certifications and degrees, and building real-world skills through CTFs and labs to what it’s like to hack AI systems, how those attacks work in practice, and where the field is heading.

People:
    SpeakerBio:  Clayton Boozell
No BIO available

Creator: DCNextGen

And just like that, it’s a wrap! While we’re sad to see the fun end, we want to give everyone one last big thank you.

Join us to say goodbye to new friends and hear about all the cool plans we have for next year. We’ll also be handing out prizes for our DCNextGen Badge CTF — you must be present to win!

People:
    SpeakerBio:  BiaSciLab
No BIO available

Creator: DCNextGen

Come pick up your DCNextGen electronic badge and swag! Get a preview of all the upcoming activities and adventures. We’ll also show you how to use your new badge in order to participate in all of our cool ctf challenges around the conference!

People:
    SpeakerBio:  BiaSciLab
No BIO available

Creator: La Villa Community

En esta charla práctica y directa al punto, se demostrará cómo una vulnerabilidad de buffer overflow puede ser explotada para comprometer entornos SCADA, evolucionando desde la ejecución de shellcode multietapa hasta el establecimiento de canales de comunicación C2 y el control remoto de sistemas industriales críticos. A través de demostraciones técnicas, se analizará cómo una vulnerabilidad aparentemente aislada puede convertirse en una cadena de ataque capaz de afectar procesos industriales, provocar interrupciones operativas e incluso derivar en escenarios de blackout.

People:
    SpeakerBio:  Fernando Mengali, Information Security Specialist

Cybersecurity researcher focused on Pentesting and AppSec, also serving as a Practical / Hands-on Speaker. He also dedicates part of his research to critical infrastructure security (ICS/SCADA), exploring the intersection between classic vulnerabilities and industrial environments. He has over 18 years of experience in offensive security and practical application exploitation.

He has participated in research and development projects focused on vulnerability exploitation and offensive security.

Specialized in 0day discovery and exploit development, he has over 135 published CVEs and is ranked in the Top 10 contributors on VulDB https://vuldb.com/users.top.

Additionally, he has published multiple exploits and technical papers publicly available https://www.exploit-db.com/?author=12136 and https://packetstorm.news/files/author/8470/1.

His work focuses on real-world vulnerability exploitation, including advanced scenarios such as memory corruption, buffer overflows, and complex environments.

He is the creator of https://yrprey.com, an educational framework that brings together more than 20 vulnerable applications based on the OWASP Top 10, used for practical training in Application Security and Offensive Security https://owasp.org/www-project-vulnerable-web-application-directory.

He is also the driving force behind https://speakfy.io, a project focused on promoting Information Security events globally.

Furthermore, he develops application security-oriented tools, such as https://leapfix.co (static code analysis) and https://fitoxs.com, a dynamic application analysis platform powered by artificial intelligence.

Creator: The Diana Initiative

While purple teaming isn’t a new concept, getting purple teaming actually started and happening consistently is tough! Maybe your company or team experienced the one purple team event, then before you know it was back to red vs blue? Or maybe your purple team just never happened? Join Allie as she discusses her experiences with this struggle – where she dealt with only taking red team reports to make the blue team better, to finally getting them to collaborate and improve both teams!

People:
    SpeakerBio:  Allie

Allie is a Senior Security Analyst with eight years of combined experience across systems administration, networking, and cybersecurity. She has previously presented work about threat hunting nation state actors targeting the open-source software ecosystem. She spends her days hunting threat actors and leading incident response investigations. Outside of work, she enjoys camping, spending time with her dog, and her kids, who rank somewhere between “favorite humans” and “active incident response scenario” depending on the day.

Creator: Payment Village

Deepfake defense is no longer just a matter of classifying media as real or fake. As attackers adopt new generators, optimize them for specific targets, and automate feedback-driven attack loops, the threat is evolving into agentic fraud. This talk presents an adversarial research approach to defending live identity-verification systems: continuously simulating realistic attacks, generating and evaluating domain-specific synthetic media, turning detector failures into new evaluation data, monitoring emerging tools, and rapidly adapting to unseen generators without overfitting to the latest attack. Ultimately, effective defense requires treating deepfake detection not as a static model-building task, but as a continuously evolving adversarial process in which success is measured by resilience against the next attack, not performance on the last one. We’ll extend these concepts in the village through a hands-on challenge, inviting participants to evaluate a blind synthetic-data discrimination task, which illustrates the practical challenges of evaluating increasingly capable adversarials.

People:
    SpeakerBio:  Efim Boieru, Senior Manager of ML Engineering, Incode Technologies

10+ years in applied AI, computer vision and biometric security, focused on face liveness and deepfake detection; previously in ML research/engineering roles at Huawei and Bosch.

Creator: OWASP Foundation

AI systems face threats that traditional application security standards weren’t built to address. This includes prompt injection, training data poisoning, model extraction, agentic autonomy risks, and more. The OWASP AI Security Verification Standard (AISVS) provides 400+ testable requirements across 14 chapters, covering everything from input validation and model lifecycle management to MCP protocol security and autonomous agent controls. This lightning talk introduces the standard’s structure, its three verification levels, and how security teams can use it today to assess and harden AI-powered applications. We’ll show where AISVS fits alongside existing frameworks like ASVS, NIST AI RMF, and ISO 42001 and where it deliberately doesn’t overlap.

Links:
    Website – https://owasp.org/www-project-artificial-intelligence-security-verification-standard-aisvs-docs/

People:
    SpeakerBio:  Jim Manico, Founder at Manicode Security

Jim Manico is the founder of Manicode Security, where he specializes in training software developers on secure coding and security engineering. He is actively involved in multiple ventures, serving as an investor/advisor for companies like 10Security, MergeBase, Nucleus Security, KSOC, and Inspectiv, among others. Jim is a recognized speaker, focusing on secure software practices, and holds a distinguished position as a member of the Java Champion community. He is also the esteemed author of “Iron-Clad Java: Building Secure Web Applications” published by Oracle Press.

Additionally, Jim generously volunteers for the OWASP foundation, co-leading key projects such as the OWASP Application Security Verification Standard and the OWASP Cheatsheet Series.

Creator: Middle Easterns & Africans in Cyber Security (MEACS)

Every week brings a new agentic AI attack and a new point solution that claims to stop it. None of them stop all of it, because agents do not have one attack surface, they have many, spread across the skills they load, the tools they call, the data they touch, and the humans who approve them. This talk introduces the AISECA Tiered Control Framework, an open, community-built model for defending agentic workflows the way we defend everything else that matters: in depth, in layers, with no single control carrying the whole load. We walk the framework end to end, then go deep on one control to show how it works in practice, what it catches, what it misses, and how it composes with the controls around it. You will leave with a shared vocabulary for agentic risk and a practical way to find the gaps in your own stack. Built in the open by practitioners, free to adopt, and yours to extend.

People:
    SpeakerBio:  Amber Bennoui
No BIO available
SpeakerBio:  Karol Piekarski, DevOps Engineer

Karol Piekarski is a Lead DevOps Engineer working across cloud infrastructure, zero-trust architecture, and AI and agentic security for systems that serve hundreds of millions of consumers. His research focuses on the security of large language models and autonomous agents, with an emphasis on practical red-teaming and defensive tooling that teams can actually operationalize rather than shelve.

His empirical work on abliteration and agentic framing reframes where alignment actually breaks down in agent pipelines: agentic framing alone can collapse a model’s baseline safety, while abliteration matters mostly for the hardest content and is highly architecture-dependent. In 2026 he co-presented “Move Fast, Stay Secure: Enterprise AI Agent Security in Practice” at the Databricks Data + AI Summit, and spoke at SCaLE 23x on open source red-teaming for LLMs. He was one of only two external contributors to the Databricks Agentic AI Security Framework (DASF v3.0).

Karol is the creator of Sundew.sh, an open source, MCP-native AI agent honeypot platform that uses behavioral fingerprinting and a persona engine for anti-fingerprinting, now adopted by external research teams studying agent behavior in the wild. He was selected as a Wiz MVP last year and this year received Wiz’s Thought Leader award, and he is active in the AISECA Working Group, where he co-owns agentic security risk definitions.

He holds the CCSP, CKA, four AWS specialty certifications along with DataDog and Tines, and he regularly judges and mentors at hackathons across Southern California.

Creator: La Villa Community

Hay muchas herramientas para hackear RF y mucho conocimiento regado en talks, repos y la cabeza de la gente. Lo que no hay es un mapa. Cuando te paras frente a un dispositivo o señal, ¿por dónde empiezas? ¿y cómo sabes qué se te escapó?

Esta charla cubre tres investigaciones reales, cada una en una tecnologia distinta.

Primero BLE, sniffeando advertising packets terminé auditando una sala y encontré 6 de 85 dispositivos completamente controlables, incluso con LE Secure Connections habilitado.

Después LoRaWAN, capturé 51,304 frames en US915. El 89.3% eran JoinRequests: una red entera fallando en unirse, visible sin transmitir un solo paquete, red que no podia encontrar con un SDR.

Y al final LTE — encontrar la celda correcta en México, entender el OFDM, y extraer metadatos en pasivo.

Tres casos, tres analisis distintos. Pero siempre sobre las mismas capas: espectro → señal → enlace → cripto → ataque.

Ya seguía una metodología, solo no la había nombrado.

Así nació RFSAM (Radio Frequency Security Assessment Methodology): una referencia abierta, estructurada, que organiza lo que OSSTMM, BSAM y la comunidad SDR ya construyeron, en algo que puedes navegar. No pretende inventar nada, pretende ser un mapa.

https://electroniccats.github.io/RFSAM/

People:
    SpeakerBio:  Eduardo Contreras, CTO

Electronics Engineer, passionate about technological development

and science, co-founder and CTO of the Electronic Cats company which has developed different embedded systems already in production, focused from education to security related devices, embedded programer, has developed open source libraries for communities as well as open hardware, with the goal to design and implement global impact electronics.

Creator: Malware Village

In recent years, ransomware has become one of the most prolific forms of cybercrime with financial gain as primary motive. The problem keeps getting bigger, with a new operation seeing the light almost every month. The Dutch National Police is often a key player in large-scale ransomware investigations. Operation Cronos is a prime example, where law enforcement took down infrastructure of the infamous LockBit group in 2024.

Today, many ransomware groups operate RaaS (ransomware-as-a-service) models. They provide the ransomware and a platform to extort victims as a service, and affiliated hacker groups carry out the actual attacks. The platform is often run from a VPS (virtual private server), and sometimes, this server is in the Netherlands. Dutch law enforcement seizes those servers and extracts their content for investigation. This happened to the Playboy ransomware in late 2024.

A chain of various tools is used to provide a ready-to-use ransomware package to affiliates of a RaaS program. Several modern ransomware operations even support CPU architectures and operating systems other than x86-64 and Windows. New cryptographic keys are generated for every victim, ransomware parameters are configured, and a builder creates the final package to be used by the affiliate. In this talk, we will reveal how we seized the Playboy ransomware servers, dive into its toolchain, and look at how executables were prepared to breach victims.

People:
    SpeakerBio:  Gijs Rijnders, Malware & CTI Specialist

Gijs is a cyber threat intelligence analyst and malware reverse engineer at the Dutch National Police where he defends the Police organization from cyber attacks. He previously worked at the CERT of Tesorion, a Dutch cybersecurity company, where he reverse engineered various ransomware families and published decryption tools to the NoMoreRansom initiative to help victims recover from attacks.

Creator: Blacks In Cyber Village

Dive into the future of cybersecurity with a session that challenges conventional thinking. This talk explores the concept of ‘DeNISTifying Technology,’ guiding attendees through paradigm shifts towards quantum encryption, advanced post-cryptography methods, and the evolving landscape of digital forensics. Discover innovative approaches and critical insights into securing our digital world beyond current standards. All experience levels are welcome to explore these cutting-edge topics.

People:
    SpeakerBio:  Aisha Berry, University of Maryland Global Campus

Aisha Berry is a cybersecurity professional, PhD candidate, and experienced nurse with over 15 years in healthcare. They hold a Bachelor’s in Cybersecurity and Computer Networking with a focus on digital forensics, and a Master’s of Science in Digital Forensics and Cyber Investigations. Currently pursuing a doctoral program with a focus on healthcare, Aisha delves into the psychological component of technology, emphasizing the mindset of attackers in cybersecurity strategy.

Creator: La Villa Community

¿De verdad puede la IA encontrar vulnerabilidades reales y nuevas a gran escala? La seguridad de aplicaciones es nuestra disciplina, y somos una CNA de tipo investigador, entre las primeras del CNA Scorecard. Montamos una cadena de producción y la soltamos sobre 36 proyectos de código abierto. Hasta ahora, y contando: 539 hallazgos revisados, depurados hasta 8 CVEs públicos y 30 candidatos a CVE.

La charla recorre la cadena completa: cómo entrenamos agentes que recorren el código, cómo un modelo de machine learning detecta los fragmentos sospechosos, cómo los hackers y researchers validan cada candidato y qué números deja cada etapa del filtrado.

Nuestra meta: acortar el tiempo de detección a escala sin sacrificar exactitud, para que la ventana de exposición (detección + remediación) sea menor que el tiempo de explotación. Te llevas un método replicable para convertir LLMs impredecibles en una fábrica seria de CVEs, con las métricas de precisión que la sostienen.

People:
    SpeakerBio:  Simon Correa, Fluid Attacks, Head of Research

Simón Correa, Head of Research en Fluid Attacks, es responsable de coordinar el CNA de la compañía, realizar investigaciones técnicas de ciberseguridad y liderar la detección de vulnerabilidades en software open source de terceros, gestionando todo el ciclo de vida de los CVEs desde su descubrimiento hasta su publicación. Además, coordina y participa en el equipo interno de CTFs, analizando el rendimiento y diseñando estrategias de mejora para futuras competencias. En su rol, mantiene la interlocución con MITRE y NIST, prepara publicaciones académicas y ponencias para conferencias especializadas, y organiza meetups técnicos que fortalecen el posicionamiento de Fluid Attacks en la comunidad de seguridad.

SpeakerBio:  Michael Rivera, Chief Data & AI Officer | M.Sc. in Analytics

Michael Rivera is the Chief Data & AI Officer at Fluid Attacks, where he leads initiatives that combine advanced data analytics and artificial intelligence to strengthen software quality and security. His work in recent years has focused on developing methods that reduce the time required to detect vulnerabilities and on ensuring that AI-driven systems are built under the security-by-design approach.

Michael combines an academic background in economics with his Master of Science in Analytics at the Georgia Institute of Technology, enabling him to bring an uncommon interdisciplinary perspective to the field of cybersecurity. With years of experience bridging data science and AppSec practice, he offers both technical depth and strategic insight, making him uniquely positioned to help organizations adopt rigorous frameworks for evaluating security solutions.

Creator: Red Team Village

DFMI is a cross-platform, open-source offensive toolkit that hijacks & abuses the Windows Installer’s own execution engine to detonate arbitrary payloads during software installation, with zero files written to disk and zero evidence left behind. And this can be achieved without even corrupting the Authenticode of the binary.

The attack surface is the CustomAction table embedded in every MSI database — a small structure that Windows Installer processes unconditionally at install time. DFMI injects a deferred CustomAction (property-based EXE invocation) — firing right before the modification in system files; the payload executes before a single legitimate byte touches the filesystem.

Right now, DFMI provides 3 different methods for abusing MSI files: – Inject: Simply injects a CustomAction (CA) into an existing MSI package. – Rogue MST: Generates an MSI Transform File (.mst) and injects into legitimate “.msi” file without corrupting it’s Authenticode. – Stub: Creates a malicious MSI file from scratch.

What makes DFMI particularly difficult to eradicate is that it exploits no vulnerability; it’s a feature. The CustomAction mechanism is a 25-year-old Windows feature, documented by Microsoft, used by virtually every enterprise software package and trusted implicitly by the OS.

DFMI simply turns that trust sideways — using the Installer engine as its own payload executor.

https://github.com/ccelikanil/DFMI

People:
    SpeakerBio:  Anıl Çelik

Computer Engineer & been working as a Red Teamer for the past ~7 years. Previously did presentations at DEFCON 33 Demo Labs, DEFCON 33 Red Team Village & Black Hat USA Arsenal 2025. Currently holding 6 CVEs, OSCP & OSWP. Interests: Windows Internals & AD Security

Creator: Blue Team Village

Forensics 101 is a fast-paced, story-driven primer on digital investigations—perfect for incident-response newcomers and seasoned blue-teamers alike. In under 45 minutes we trace the full evidence lifecycle: imaging disks and cloud buckets, teasing meaning from volatile memory, and translating binary breadcrumbs into courtroom-ready narratives. First, we demystify the tools and techniques behind solid acquisition and analysis; next, we zoom out to the real-world team dynamics, reporting pitfalls, and career launchpads that turn analysts into trusted advisors. Expect war stories, practical tips you can use on Monday, and zero certification flexing—just the art, science, and occasional 2 a.m. data-center comedy of modern digital forensics.

People:
    SpeakerBio:  Sarthak Taneja

Started from Offensive Security and ended up in Defense. Wearing Purple hat most of the days.

Creator: Ham Radio Village

Digital communication has become one of the fastest-growing areas of amateur radio. By combining traditional radio equipment with computers and software, operators can send text messages, telemetry, position data, and even digitally encoded voice across the airwaves.

This presentation introduces the fundamentals of amateur radio digital modes and how they work. We’ll begin by explaining the difference between analog and digital signals and how computers convert data into audio that radios can transmit.

The talk then explores several widely used digital modes, including FT8 and WSPR for weak-signal communication, JS8Call for keyboard-to-keyboard messaging, and APRS for real-time position tracking and messaging. We’ll also look at newer developments such as LoRa-based APRS experimentation and open-source digital voice technologies like FreeDV and M17.

Finally, we’ll discuss practical ways to get started with digital radio using affordable hardware, including simple USB radio interfaces, beginner-friendly radios, and lightweight portable setups.

Attendees will leave with a clear understanding of how digital radio works and several easy paths to begin experimenting with digital modes themselves.

Modern amateur radio is no longer just voice communication. Today, radios routinely exchange digital data, text messages, telemetry, and even machine-learning encoded voice signals.

This talk introduces the world of amateur radio digital modes and explains how computers and radios work together to communicate using sound. We’ll explore some of the most popular digital modes used by operators today, including FT8, JS8Call, WSPR, APRS, and several emerging digital voice technologies.

Attendees will learn what digital modes are, how they differ from traditional analog radio, and why they have become so popular. The session will also highlight new developments in the digital radio ecosystem, including open-source digital voice systems and long-range LoRa APRS experimentation.

If you’ve ever wondered what those strange buzzing sounds coming from a radio actually mean—or how to start experimenting with digital radio yourself—this talk will show you how to get started.

People:
    SpeakerBio:  Jon Marler (K4CHN)

Jon Marler is a cybersecurity product director, hacker, and RF experimenter who enjoys building strange things that connect computers, radios, and networks together. By day he leads security products at VikingCloud focused on protecting global payment infrastructure. By night he experiments with mesh radio networks, hardware hacking, and unconventional communication systems. Jon is particularly interested in how radio and decentralized networks can complement modern security and resilience strategies.

Creator: Red Team Village

Command and control beacons currently reach out to the operating system to request socket resources, triggering auditing and being restricted by local firewall rules. However, it is possible to develop malware that can evade the standard process. Sparse is a tool that weaponizes this concept to evade such auditing and defensive configurations.

People:
    SpeakerBio:  Andrew Rioux

I started learning to program in middle school and found a software engineering project to work on throughout highschool. I attended Liberty University to study Software Engineering and Cybersecurity at the undergraduate level, learning to develop software that works in adversarial contexts. I have recently finished my master’s degree in Cybersecurity, where I was able to further practice developing software in adversarial contexts and develop version 2 of Sparse

Creator: OSINT For Good Community

OSINT, social media, drones, AI, and more… in this session, we’ll learn how these innovations are applied to make life-saving decisions before, during, and after a disaster

People:
    SpeakerBio:  Matt Green, Green Emergency Management Services

Matt Green is your friendly neighborhood emergency manager, professor, and host of the state of disaster podcast. His passion is ensuring equitable, high-quality, whole community emergency management across sectors and borders.

Creator: OSINT For Good Community

OSINT, social media, drones, AI, and more… in this session, we’ll learn how these innovations are applied to make life-saving decisions before, during, and after a disaster

People:
    SpeakerBio:  Matt Green, Green Emergency Management Services

Matt Green is your friendly neighborhood emergency manager, professor, and host of the state of disaster podcast. His passion is ensuring equitable, high-quality, whole community emergency management across sectors and borders.

Creator: Voting Village

In 2022, a Dominion Voting Systems ImageCast X ballot marking device (ICX BMD) appeared for sale on eBay and was purchased by security researcher Harri Hursti. Originating in Colfax Township, MI and after being “donated by accident” to Goodwill, it debuted at Voting Village 2023 and has since been the subject of great interest and used for proof-of-concept demonstrations (PoCs) within the village. As the only known instance of a legally-obtained ICX being available for study without burdensome secrecy/pre-publication review agreements, it represents a unique opportunity for improving and updating our understanding of certified, adopted, and deployed voting systems from an engineering and security perspective (among others).

In this talk, we will discuss the hardware, software, architecture, and capabilities of the ICX BMD from a technical perspective. Significant reverse-engineering of its Democracy Suite 5.5 software has revealed much regarding its design, construction, and internal operation along with a few surprises along the way. While one part of one version of one system from one vendor is far from sufficient to draw conclusions about currently-used voting systems in-general, this talk hopes to shed light on what a currently EAC-approved voting system could be.

People:
    SpeakerBio:  Drew Springall

Drew Springall is a hacker, security researcher, and frequent Voting Village contributor with a specific interest in the technical/concrete aspects of voting system security. Since 2013, Drew has worked to understand the challenges encountered by and demonstrate the capabilities of realistic attackers should they attempt to interfere with such systems as commonly deployed. In those years, Drew has built and demonstrated many proof-of-concept attacks across various areas of voting system security including Internet Voting, voter-facing software/hardware, ballot de-anonymization, and physical protections.

Creator: Red Team Village

“Identity is the new perimeter” gets thrown around a lot in security discussions, fo a good reason. With the right privileges, cloud resources can be managed through APIs. Each provider has their own way of handling API access on resources. If your identity has the right permissions for a specific API call, you can execute that action.

The thing is, it’s hard to remember every single API call you need to get specific information or access certain resources. AWS recognized this problem and created the AWS CloudControl API, a unified API that groups different resources together so you can create, update, delete, or retrieve them in bulk. This makes managing resources much simpler since you don’t need to know which specific service, API call, or parameters to use.

But here’s the catch: when something is easy for legitimate users, it’s usually easy for attackers too. There are already tools available to use the AWS CloudControl API to pull resource information, and the technique is gaining attention.

But is it actually a good way to do stealthy enumeration? In this article, we examine what the CloudControl API is, how it simplifies resource management, how attackers can weaponize it, its limitations, and detection methods.

People:
    SpeakerBio:  Bleon “gl4ssesbo1” Proko

Bleon is an Info-sec passionate about Infrastructure Penetration Testing and Security, including Active Directory, Cloud (AWS, Azure, GCP, Digital Ocean), Hybrid Infrastructures, as well as Defense, Detection and Thread Hunting. He has presented topics related to Cloud Penetration Testing and Security in conferences like BlackHat USA, Europe and Sector, DEF CON, SANS Pentest Hackfest Hollywood and Amsterdam, as well as several BSides on USA and Europe.

His research include Nebula, a Cloud Penetration Testing Framework (https://github.com/gl4ssesbo1/Nebula) and other blogs, which you can also find on his blog (blog.pepperclipp.com). He is also the author of YetiHunter, DetentionDodger and Ransomwhen (https://github.com/Permiso-io-tools/[DetentionDodger | YetiHunter | Ransomwhen]).

He is also the author of the upcoming book “Deep Dive into Clouded Waters: An overview in Digital Ocean’s Pentest and Security” (https://leanpub.com/deep-dive-into-clouded-waters-an-overview-in-digitaloceans-pentest-and-security)

Creator: Misc

Donating bone marrow can save lives, but many don’t understand what it really involves. Find out the experience and how you can help participate in one of the coolest bio-hacks ever.

Links:
    Website – https://www.nmdp.org/

Creator: Red Team Village

Complete infostealer kill chain hands-on, executed against Windows 11 21H2 on default UAC using only open-source tooling. No zero-days, no commercial C2. Participants run every command live: HTA delivery, Meterpreter in-memory session, Chrome and Edge credential extraction via DPAPI-locked SQLite, keylogging in explorer.exe memory, UAC bypass via fodhelper, SYSTEM via Named Pipe Impersonation, and LSASS dump via Kiwi. The DPAPI_SYSTEM key recovered from LSA secrets decrypts the browser credentials captured earlier, closing the loop.

OUTLINE

Stage 1: generate HTA payload, deliver via Python HTTP server, open Meterpreter session Stage 2: extract Chrome and Edge credentials via post module, inspect DPAPI-encrypted loot Stage 3: migrate to explorer.exe, start keyscan, capture live keystrokes Stage 4: UAC bypass via fodhelper registry hijack, elevate to SYSTEM via Named Pipe Impersonation Stage 5: load Kiwi, dump NTLM hashes, SAM, LSA secrets, extract DPAPI_SYSTEM key, decrypt browser loot Detection debrief: Sysmon Event ID 1 process tree and PowerShell Script Block log review per stage

TAKEAWAYS

1. A complete credential theft playbook from HTA delivery to LSASS dump, validated against Windows 11 21H2 with Chrome 147 and Edge 147 on default UAC, ready to adapt for engagements.
2. The DPAPI close-loop: how DPAPI_SYSTEM from LSA secrets decrypts browser credentials collected earlier, giving operators a concrete credential pivot from a single user-context session.
3. A detection gap map calibrated against real Sysmon telemetry showing which events fire on default Windows 11 and which require EDR behavioral rules most organizations do not have deployed.

People:
    SpeakerBio:  Filipi Pires, Head of Technical Advocacy at SCYTHE

I’ve been working as Head of Technical Advocacy at SCYTHE, Founder & Investor at CROSS-INTEL, Advisor & Investor at Sherlockeye, BSides Porto Organizer, Red Team Village Director (DEF CON), Senior Advisor Raices Cyber Academy, Founder of Red Team Community (Brazil and LATAM), AWS Community Builder, Snyk Ambassador, Application Security Specialist and Hacking is NOT a crime Advocate. International Speaker at Security and New technologies events in many countries such as US (Black Hat & Defcon), Canada, France, Spain, Germany, Poland, Black Hat MEA – Middle-East – and others, I’ve served as University Professor in Master Degree in Portugal, Graduation and MBA courses at Brazilian colleges, in addition, I’m Creator and Instructor of the Course – Malware Attack Types with Kill Chain Methodology (PentestMagazine), PowerShell and Windows for Red Teamers(PentestMagazine) and Malware Analysis – Fundamentals (HackerSec).

Black Hat US 2025 – https://blackhat.com/us-25/arsenal/schedule/presenters.html#filipi-pires-46329 Black Hat US 2024 – https://blackhat.com/us-24/arsenal/schedule/presenters.html#filipi-pires-46329 Black Hat MEA 2025 – https://blackhatmea.com/speaker/filipi-pires-0 Black Hat MEA 2024 – https://blackhatmea.com/speaker/filipi-pires DEF CON 33 / 32 – https://sessionize.com/filipi-pires/ DEF CON – Adversary Village – https://adversaryvillage.org/adversary-events/DEFCON-33/Filipi-Pires/

Creator: IoT Village

An AI agent found a previously undisclosed vulnerability in a named vendor’s IoT product within hours of being pointed at the firmware. The practitioner who built the agent had spent 25+ years doing that work by hand. Plan R is an IoT-focused MCP server that gives AI agents direct access to real pentesting tools, structured by playbooks that encode methodology rather than scripts. The framework expanded from firmware-only analysis to a multi-domain suite covering WiFi, BLE, network protocols, and hardware interfaces, with each domain compounding the value of every other through cross-domain correlation. The centerpiece is a real engagement: a named vendor, a disclosed vulnerability, and the specific challenge of convincing a white box vendor that a finding is real when the methodology that found it is “an LLM read your code.” Attendees leave with a working blueprint for building their own AI pentesting agents, an honest account of where the approach breaks, and a direct answer to the question every experienced practitioner is quietly asking: if an AI can do this, what exactly am I bringing to an engagement? The answer is worth hearing. But the work is changing, and this community should be driving how.

People:
    SpeakerBio:  Larry Pesce
No BIO available

Creator: Maritime Hacking Village

HMI engineering tools are trusted gateways into industrial environments. Engineers use these tools to design screens, configure tags, connect to controllers, and deploy projects to real-world sites. HMIs are used across manufacturing, logistics, maritime, energy, utilities, building automation, and many other industrial sectors. Yet the engineering tools and project files used to create them are often treated as simple work utilities.

This presentation analyzes multiple vulnerabilities discovered in the HMI engineering toolchain from a supply chain perspective. It covers memory corruption during project file parsing, DLL search path hijacking, the loading of unsigned components, UI spoofing through silent font installation, and fallback to plaintext policies when secure OPC UA connections fail.

The core argument is simple: the ICS supply chain does not begin only at a vendor’s update server. Project files received from customers, templates shared by system integrators, maintenance backups, local DLLs, fonts, communication settings, and the engineering workflow of “open, drag, and deploy” are all part of the supply chain.

This presentation shows that not only PLCs and HMI runtimes, but also the tools and files used to create them, are part of the attack surface.

People:
    SpeakerBio:  Jiwoon Yoo, CYTUR Inc

CYTUR Inc. / Security Team Lead

Research Areas and Interests Maritime Cybersecurity, Satellite Security, Large Language Models, Adversarial Attacks Education and Professional Experience 2017: B.S. in Computer Science Engineering, Anyang University, Gyeonggi-do, South Korea 2023: M.S. in Convergence Security, Graduate School of Information Security, Korea University, Seoul, South Korea 2023 – Present: Security Team Lead, CYTUR Inc.

Academic Activities Symposium Participation * Nov 13–14, 2024: Participated in CFP-selected research at the IMO Maritime Cybersecurity Symposium (United Nations), London, UK

Journal Publication Sensors (May 24, 2023) — First author: Formulating Cybersecurity Requirements for Autonomous Ships Using the SQUARE Methodology https://doi.org/10.3390/s23115033

Sensors (Feb 26, 2022) — First author: Cyberattack Models for Ship Equipment Based on the MITRE ATT&CK Framework https://doi.org/10.3390/s22051860

SpeakerBio:  TaeWoo Kim, CYTUR Inc
No BIO available
SpeakerBio:  Eunji choi, CYTUR Inc
No BIO available

Creator: Game Hacking Village



People:
    SpeakerBio:  Piffd0s
No BIO available

Creator: Game Hacking Village

The PlanetWeb Internet Browser v3.0 (2000–2001) is the official disc that let you browse the web and read email on a Dreamcast. Its online features depend on a backend server that no longer exists. The dreamcast is now a networked, code-downloading appliance whose trust model rests on a domain anyone can now point wherever they want.

What We Did

This is a full unauthenticated, network-only attack. By answering DNS for the dead backend domain, we get the browser to download and run our code. From there we chain through the mail client to reach the console’s bare metal, and end by running DOOM natively.

No physical access, no modchip, no user trickery beyond opening one email. The console was built to download and trust code from a server that’s gone — so we just become that server.

The talk walks through the chain at a high level, the obstacles of debugging a 24-year-old black box, and what it takes to get a modern game engine running in the few megabytes of RAM the Dreamcast has — ending with a live demo of DOOM running on the browser.

Why It Matters / Why It’s Fun

It’s a clean case study in why abandoned online infrastructure is a security problem that outlives the product, and it’s a nostalgic hardware-hacking story with the most satisfying possible payoff: DOOM, on a Dreamcast, over the internet.

People:
    SpeakerBio:  Christopher Hernandez, Binary Enthusiast

Chris Hernandez is a security researcher specializing in vulnerability discovery, exploitation, and large scale reverse engineering using IDA Pro. A Pwn2Own competitor in the ICS/SCADA and embedded systems categories, he actively collaborates with the reverse engineering community to solve binary analysis challenges.

—

Chris Hernandez is the founder of Adversary Consulting Group and an offensive security researcher with more than a decade of experience in vulnerability research, reverse engineering, and exploit development. He holds the OSEE certification and has competed at Pwn2Own in the IoT and ICS/SCADA categories.

Creator: Data Duplication Village

For over 14 years, Backblaze has been collecting the failure rates of hard drives in our data centers. But, what does that even mean? How do you define a failure, and how does that definition define or obscure the realities of drive performance?

People:
    SpeakerBio:  Stephanie Doyle, Sr. Manager and Keeper of Stats at Backblaze

Stephanie Doyle is a Sr. Manager at Backblaze and known as the Keeper of Stats. She oversees the various first party data reports including Drive Stats, which reports on the failure rates of hard drives in Backblaze data centers; Network Stats, which reports on network traffic into and out of Backblaze’s network; and Performance Stats, which publishes quarterly testing on cloud storage benchmarks. She specializes in taking complex topics and writing relatable, engaging, and user-friendly content. You can most often find her reading in public places, and can connect with her on LinkedIn.

Creator: Radio Frequency Village

The low-altitude airspace has an RF problem. The FAA mandated Remote ID as the foundational identification standard for drone operations across the United States – a broadcast protocol running on Bluetooth and WiFi that any phone is supposed to receive. The problem is that it was designed for individual accountability, not area surveillance, has a practical detection range measured in hundreds of meters, is completely unencrypted and unauthenticated, and is trivially spoofable with an ESP8266 and three dollars in hardware. Meanwhile the proprietary protocol it largely replaced – DJI’s OcuSync Drone ID – was detectable at tens of kilometers, is now encrypted on newer hardware, and its primary detection product was discontinued. The gap between what practitioners think Remote ID provides and what it actually provides is a fundamental RF problem with direct public safety consequences.

This talk lives in the RF layer. We will cover what drone broadcasts actually look like on a receiver – Remote ID over BLE advertising channels 37/38/39 and WiFi beacon frames on channels 6 and 149, DJI Drone ID embedded in OcuSync, and what non-broadcasting drones look like (nothing). We will walk through the full detection stack from a $50 COTS build (Ubiquiti Bullet M2, OpenWRT, tcpdump, Wireshark, OpenDroneID dissector) through man-portable dedicated receivers (DroneTag Rider), distributed mobile detection via ATAK plugins (Drone Hone), and enterprise fixed-site systems – and we will show real sensor performance data from multi-sensor deployments at major federal security operations. Key finding: 95.5% of unique drones detected across a SEAR 1 National Special Security Event were seen by only a single sensor. The multi-sensor fusion problem is unsolved at urban scale, and the urban RF environment – path loss, noise floor, building shadows — is why. We will then walk through the spoofing ecosystem in depth. Ghost swarm attacks (jjshoots/RemoteIDSpoofer – 16 fake drones, web UI, $3 ESP8266), pilot location spoofing (brinuk/Remote-ID-Drone-and-Pilot-Spoofer – sends ground teams to empty parking lots), DJI Drone ID spoofing (DJISDKUser/ESP8266_DJI_DroneID_Throwie – Kevin Finisterre, taped to a fence, runs forever), and protocol-level implementations (cyber-defence-campus/droneRemoteIDSpoofer – Python/Scapy, ASTM F3411 plus DJI format, Swiss Cyber Defence Campus provenance). All public, all GitHub, all under $5 hardware. We will cover the attack taxonomy: evasion, flooding, identity spoofing, and operator location spoofing – and what each one means operationally for the detection stack.

The legal layer is real and the audience should know it: 6 USC 124n authorizes four federal agencies to take action against drones. Nobody else can jam, spoof, or kinetically defeat without federal authority. The possession/testing/broadcasting distinction matters and we will be clear about it.

We close with the governance gap: Remote ID’s authentication-free design was a deliberate policy choice. The FAA Reauthorization Act of 2024 directed a review of whether cryptographic signing could be added. That review is ongoing. The people in this room build the tools that inform these conversations – this is the right audience to have opinions about what the answer should be.

No policy background required. If you know what a waterfall is you are the target audience.

People:
    SpeakerBio:  Greg Albrecht
No BIO available

Creator: Aerospace Village

Counter-Unmanned Aircraft Systems (C-UAS) is a domain spanning RF engineering, federal law, operational security, and public policy, and it’s being built right now by people working without a complete picture. Most security practitioners and nearly all civilians don’t know how drone detection works, what the legal framework for defeating drones is, who’s authorized to do what and why, or how badly the current technical stack can be defeated with a microcontroller. Drawing on direct operational experience at SEAR 1 National Special Security Events (the highest domestic security classification in the United States), this presentation walks through the complete C-UAS stack from first principles, grounded in real operational data rather than vendor marketing.

People:
    SpeakerBio:  Greg Albrecht
No BIO available

Creator: Bug Bounty Village

At HackerOne, the same community that submits vulnerability reports also tests the platform they use to report them. Every report becomes a live stress test of our product, workflows, assumptions, and newly shipped features. In this talk, a Senior Triage Lead and a Product Security Engineer share what we learned from running HackerOne’s own bug bounty program while shipping GraphQL features, AI-assisted tooling, disclosure workflows, and platform-scale infrastructure changes. Using three real disclosed reports, we walk through how seemingly small bugs escalated into platform-wide security lessons: 1. An Elasticsearch query parameter that enabled metadata enumeration through raw script execution. 2. A PDF export feature that unintentionally exposed internal triage activity. 3. An AI agent that exposed non-public report metadata because a researcher asked the right question. We will show how reports move from submission to validation, severity debates, engineering response, remediation, retesting, and disclosure. We will also discuss how we use our own program as a testing ground for AI-assisted triage, automated validation workflows, and disclosure policies before rolling changes out more broadly. This is not a “how to run a bug bounty program” talk. It is a behind-the-scenes look at what happens when hackers continuously attack the bug bounty platform itself and how that pressure forces rapid security evolution. Attendees will leave with practical lessons on: 1. building tighter triage-to-engineering feedback loops, 2. handling modern attack surfaces like GraphQL and AI agents, 3. scaling remediation without losing researcher trust, 4. and turning bug bounty programs into product improvement engines instead of passive inboxes. All case studies are based on disclosed HackerOne reports. No customer data was accessed or exposed.

People:
    SpeakerBio:  Shrimant Subhash More, Senior Security Analyst, HackerOne

Shrimant Subhash More is a Senior Product Security Analyst at HackerOne with over 8 years of experience in offensive security, penetration testing, and vulnerability management. His expertise spans web, API, mobile (Android and iOS), and AI/LLM security, with more than 300 security assessments conducted across diverse industry sectors. At HackerOne, Shrimant leads and supports vulnerability triage operations, validates security reports submitted by researchers, handles escalations, mentors analysts, and collaborates with global teams to improve security outcomes and triage efficiency. He is passionate about offensive security, product security operations, security automation, and helping organizations build resilient products. Beyond his professional role, Shrimant actively contributes to the cybersecurity community as a HackerOne Community Brand Ambassador for India West (Pune), where he organizes meetups, conducts workshops, mentors aspiring researchers, and promotes responsible disclosure and bug bounty programs. He is the assignee of CVE-2020-35296 and holds multiple industry certifications across security, cloud, and AI domains.

SpeakerBio:  Martzen Haagsma, Security Engineer, HackerOne

Martzen Haagsma is a Product Security Engineer at HackerOne with a passion for building secure systems through collaboration, automation, and continuous learning. With experience spanning security testing, DevOps engineering, and technical leadership, Martzen focuses on helping organizations identify vulnerabilities, improve security processes, and strengthen their overall security posture. Known for a solution-oriented mindset and a strong belief in the power of teamwork, Martzen enjoys connecting people, sharing knowledge, and turning complex security challenges into practical outcomes. Prior to joining HackerOne, Martzen worked across both public and private sectors in roles involving security testing, agile quality engineering, and engineering leadership. Outside of work, Martzen is an active technologist with interests ranging from security and software development to automation, IoT, and 3D printing. Through writing, mentoring, and community engagement, Martzen advocates for curiosity, collaboration, and making technology more secure and accessible for everyone.

Creator: La Villa Community

Esta investigación analiza la evolución reciente de los troyanos bancarios brasileños y su expansión internacional, así como las acciones de las autoridades para frenarlos. Se examinan campañas actuales impulsadas por ciberdelincuentes latinoamericanos que no solo perfeccionan el malware bancario tradicional, sino que también incorporan fraudes móviles basados en tecnología NFC.

People:
    SpeakerBio:  Josep Albors, Head of Awarenes & Research at Ontinet.com (ESET Spain)

Josep Albors is the Head of Awareness & Research at Ontinet.com (ESET Spain). He’s a security expert with more than 21 years working in cybersecurity and specialized in security awareness. He is also the editor at the company’s blog and one of the experts writing at several other publications related with the IT security world in Spain.

He has been a speaker at some of the most important security conferences in Spain, besides collaborating with initiatives such as X1RedMasSegura, that wants to raise awareness among users so they can use Internet and technology in a safe way. Included in Ontinet’s social responsability, Josep also does awareness and cybersecurity presentations in schools and universities. He’s also a teacher in cyber security expert courses at several Spanish Universities and participates in several conferences organized by several spanish universities and Spain’s national Institute of Cyber Security. He also participated as speaker at AVAR conference in Osaka in 2019, Caro Workshop 2023 in Bochum (Germany), FIRST 2024 (Fukuoka), Virus Bulletin 2024 (Dublín), Defcon Malware Village (2025), JSAC 2025 (Tokyo) and CARO 2026 (Innsbruck).

Josep has also collaborated with the Spanish Guardia Civil, Spanish National Police and the Spanish Army, teaching their units on how to fight cybercrime and with cyber intelligence training, contributing with his experience in analyzing cybercrime and malware.

Creator: Voting Village

This talk will give an overview of the technology used in US elections, the security vulnerabilities that can compromise elections, and approaches for making elections more secure and robust.

People:
    SpeakerBio:  Matt Blaze

Matt Blaze is the McDevitt chair in Computer Science and Law at Georgetown University, where he studies problems at the intersection of technology and public policy. Election systems and voting technology are central focuses of his research. He led teams as part of the California TTBR and Ohio EVEREST studies that each found deep and fundamental weaknesses in different election technologies used by those states and the rest of the US. He has testified on technical risks in elections before the US Congress and other bodies numerous times. Blaze is a co-founder of the Voting Village and president of the Election Integrity Foundation.

Creator: Voting Village

As Dan Wallach once said, an election must convince the losers that they lost fair and square. To do this, election administrators need to do more than merely allowing observers from all parties to watch, we must ensure that observers can understand what they are seeing. Watching the back of an election administrator doing something with a computer or even a with a stack of papers is not useful. Anecdotal reports by observers can give a feel for how an election was conducted, but systematic and well documented reporting is far more powerful. Election observation is difficult with simple in-person vote-for one contests; vote by mail, Internet voting and instant runoff voting all pose serious problems for election observers. We must demand electoral procedures that are easy to understand, and we must demand public documentation of those procedures. This talk will focus on successes and failures in election observation and suggest directions for future improvement.

People:
    SpeakerBio:  Douglas W. Jones

Douglas W. Jones has been involved with election technology since he was appointed to the Iowa Board of Examiners for Voting Machines and Electronic Voting Systems in 1994. Since then, contributed to the 2002 FEC Voting System Standards, served on the Election Assistance Commission’s Technical Guidelines Development Committee, tested voting equipment in Miami and Phoenix, and observed elections in Kazakhstan and the Netherlands. He has testified about voting technology before the House Science Committee, the Federal Election Commission, and in numerous court cases. He was a principal investigator in the NSF funded ACCURATE project, and with Barbara Simons, co-authored the book Broken (CSLI Press, 2012).

He got his PhD in Computer Science from the University of Illinois in 1980, and was on the CS University of Iowa CS faculty until 2021. These days, when he is not answering questions about elections, he is working with a small group of students to restore a 1965-vintage PDP-8 computer.

Creator: Policy @ DEF CON

Election security is often framed as a technology problem. In reality, it’s a governance problem operating inside a highly stressed public system. This session brings together leaders who were inside that system during the most scrutinized election cycles in modern U.S. history, moderated by an elections policy expert who has shaped how platforms handle political content at global scale. Tim Harper, Project Lead for Elections and Democracy at the Center for Technology and Democracy and former Content Policy Associate Manager for Political Advertising at Meta, will moderate. His work focuses on combating election disinformation, strengthening election integrity, and building public trust. Lester Godsey served as CISO for Maricopa County during the 2020 and 2024 presidential elections, leading cybersecurity in a jurisdiction under intense national scrutiny and multiple adversaries. Bill Gates served on the Maricopa County Board of Supervisors during that same period, overseeing election administration amid unprecedented political pressure, misinformation and reputational and physical threats. He now directs the Mechanics for Democracy Laboratory at Arizona State University. Together, they bring a rare perspective: cybersecurity, governance, and systems engineering at the front lines of democracy.

People:
    SpeakerBio:  Lester Godsey, Arizona State University

Lester Godsey serves as the Chief Information Security Officer (CISO) for Arizona State University, the largest public university by enrollment in the United States. His appointment follows a distinguished five-year tenure as CISO for Maricopa County, where he skillfully managed information security during one of the most contentious election periods in the county’s modern history. With over 30 years of public-sector IT experience, Lester is a respected voice in the field, having presented at local, state, and national levels on a diverse range of topics including telecommunications, data management, and cybersecurity. A lifelong learner and educator, Lester holds multiple industry certifications including PMP, CISM, and CISSP. He has shared his expertise as a collegiate instructor for almost fifteen years, teaching cybersecurity, technology, and project management courses. Lester is also a published author and actively contributes to the cybersecurity community. His academic background uniquely blends the arts and technology, holding both a Bachelor of Arts in Music and a Master of Science in Technology from Arizona State University. This multidisciplinary foundation informs his innovative approach to information security leadership.

SpeakerBio:  William Gates, Arizona State University

Bill Gates is Executive Director of the ASU Mechanics of Democracy Laboratory (MODL) and Professor of Practice at Arizona State University’s Watts College of Public Service and Community Solutions. He leads MODL’s efforts to professionalize election administration through practitioner-focused training, transparency initiatives, and the integration of emerging technologies. In 2025, he launched the AI + Elections Clinic to equip election officials nationwide with the tools and expertise to responsibly harness artificial intelligence in support of voter confidence and election integrity.

Bill previously served two terms on the Maricopa County Board of Supervisors (2017–2024), overseeing elections in one of the nation’s largest voting jurisdictions during the highly scrutinized 2020 and 2022 election cycles. He also served on the Phoenix City Council (2009–2016), including as Vice Mayor.

Bill is the recipient of various awards, commending his years of leadership, service, and defense of democracy including The American Bar Association’s Unsung Hero of Democracy Award in 2024, The Truman Foundation’s Joseph E. Stevens Public Service Award in 2022, The Arizona Republic’s 2021 Arizonan of the Year, and The Phoenix Business Journal’s Forty Under 40 in 2010.

SpeakerBio:  Tim Harper, Center for Technology and Democracy

Tim Harper is Project Lead for Elections and Democracy, where he leads efforts to combat election disinformation, support secure voting technologies, and strengthen public trust in elections. An elections policy expert, Tim has driven policy and advocacy across industry and civil society. Most recently, he served as Content Policy Associate Manager for Political Advertising at Meta, where he developed and implemented global policies governing electoral, political, and social issue advertising. His work focused on transparency, preventing foreign interference, and countering election delegitimization at scale. Previously, Tim was Senior Elections Policy Analyst at the Bipartisan Policy Center, where he led election administration policy development and federal and state advocacy efforts. He convened a bipartisan task force of election officials to inform policymaking and led major data-driven research initiatives, including the largest national study on voting line disparities. He began his career supporting election management bodies in the Middle East and North Africa with the International Foundation for Electoral Systems. He graduated magna cum laude from Gonzaga University with degrees in Political Science and History.

Creator: Adversary Village

Modern intrusions often start with identity, not malware. Adversaries abuse IdP drift, device code flows, stale sessions, weak conditional access, helpdesk processes, SaaS integrations, and cloud role mappings. This talk presents a safe emulation framework for identity-first threat actors across AD, Entra ID, Okta-like workflows, and cloud control planes. The lab provides ATT&CK-aligned playbooks that simulate identity compromise, IdP pivoting, session abuse, and SaaS access without stealing real credentials. The focus is measurable purple-team validation: expected telemetry, detection hypotheses, failure points, and repeatable scoring.

People:
    SpeakerBio:  Samet Can Tasci, Senior Linux Systems Engineer

Samet Can Tasci is a Senior Linux Systems Engineer and security researcher with experience across enterprise infrastructure, cloud and hybrid environments, automation, and defensive security validation.

His research interests include web defense behavior, WAF normalization gaps, parser inconsistencies, adversary-informed testing, and practical tooling for security teams.

He is the creator of WaffleX, a research prototype selected for Black Hat USA Arsenal, which focuses on semantic consistency analysis, enforcement drift, and family-level request-variant testing across modern web application defense stacks.

SpeakerBio:  Mehmet Önder Key, Cyber Security Consultant

Önder Key is a cybersecurity consultant specializing in critical infrastructure security, zero-day vulnerability analysis, and offensive security. He has advised organizations in high-security sectors such as defense, aerospace, and finance, with hands-on experience in both red teaming and strategic security engineering. His work has been featured across numerous countries and platforms, contributing to the discovery of systemic vulnerabilities. Currently, he provides consultancy to TurkNet and continues to advance the global offensive security ecosystem by challenging traditional approaches to cybersecurity.

Creator: Policy @ DEF CON

The Common Vulnerabilities and Exposures (CVE) Program has become foundational to global cybersecurity operations. However, as vulnerability discovery accelerates and regulatory frameworks emerge, CVE is no longer just a technical reference system — it is evolving into critical infrastructure at the intersection of security and public policy. This talk presents ENISA’s operational and policy role in strengthening the global CVE ecosystem, including its designation as a CVE Numbering Authority and Root CNA, its development of the EU Vulnerability Database (EUVD), and its mandate under the Cyber Resilience Act to build a Single Reporting Platform for vulnerability coordination across Europe. Moreover, the talk will explain how ENISA plans to operationalize the collected data to strengthen the overall security ecosystem. The session will explore how policy decision at EU level impact vulnerability disclosure, coordination, and data quality — and how Europe is scaling its role without fragmenting the global system. The goal is to provide a practical, but both strategic and operational perspective on how the EU’s increased role in this area will ensure a more resilient and interoperable vulnerability ecosystem. (This talk will develop both talks held by ENISA at VulCon 2026)

People:
    SpeakerBio:  Nuno Rodrigues Carvalho, ENISA – The European Union Agency for Cybersecurity

Nuno Rodrigues Carvalho currently serves as Head of Sector of the Incident & Vulnerabilities Services within the Operations and Situational Awareness Unit (OSA) of the European Union Agency for Cybersecurity (ENISA). He leads the department in ENISA dealing with the development of Vulnerability Services (including the European Union Vulnerability Database), of the Single Reporting Platform of the Cyber Resilience Act and the Cybersecurity Incident Reporting and Analysis System (CIRAS) and the reports stemming from the NIS 2 legislation related to that system. Before joining ENISA as a Senior Threat and Vulnerability Analyst, he developed more than 15 years of experience in strategic, tactical, and operational analysis and situational awareness through roles at both national and international levels. Previously, he also worked at the European Parliament and in the banking sector.

SpeakerBio:  Razvan Gavrila, ENISA – The European Union Agency for Cybersecurity

Razvan Gavrila is a seasoned cybersecurity professional with over 15 years of experience across national and EU institutions. In his current role, he oversees ENISA’s work on the implementation of the EU Cyber Resilience Act and leads initiatives in product and security engineering, including the cybersecurity of AI systems. Prior to his appointment as Head of Sector for Market, Technology, and Product Security in January 2025, he served as ENISA’s lead Cyber Threat Intelligence (CTI) analyst. He holds several industry-recognized certifications and a Master of Science in Computer and Information Security

Creator: Red Team Village

ATMs in the end are a computer with windows

I will talk about EDRs and how parent-child process inheritance works

The Windows boot process, known as the boot process, is divided into several phases as we have already seen: PreBoot (initialization of BIOS/UEFI firmware and POST), Boot Manager (loading of bootmgr or bootmgfw.efi), OS Loader (execution of winload.exe to load the kernel), and Kernel Initialization (loading of ntoskrnl.exe and hal.dll, initialization of drivers and services). Finally, winlogon.exe is started to present the user interface. This process enables the transition from a powered-off state to a functional operating system through a hierarchical and verified sequence.

Regarding kernel-based EDRs (Endpoint Detection and Response), they operate in kernel mode, the most privileged level of the system (some also operate at the UEFI level), which allows them to monitor low-level activities such as system calls, memory manipulation, driver loading, and file access. By running in this mode, EDRs can detect and block advanced threats such as rootkits or persistent malware that attempt to evade user-mode protections. Their deep integration with the kernel enables real-time monitoring, behavioral analysis, and immediate response to suspicious activities during and after the boot process. EDRs are typically located in the kernel, but as mentioned, some also operate at the UEFI level. If we recall how Windows works, beneath UEFI lies SMM (System Management Mode), making this an even more complex layer to analyze and operate within.

This talk is intended for individuals interested not only in ATMs and simple cash dispensing, but in going beyond that—focusing on deeper analysis and the full experience gained when confronting these systems.

More information about the attack and the theory behind it: https://h0km4.com/posts/DMA-Introduction/ https://h0km4.com/posts/telemetria-bypass/

People:
    SpeakerBio:  Arnold Jared Morales Yepez, Senior Team Lead, Grupo Salinas

Self-taught in computer security since age 12; holds a degree in Computer Forensics and Cybersecurity and is pursuing a Master’s in AI and Cybersecurity.

—

Desde los 12 años, me he dedicado a la informática con un enfoque especial en la seguridad. Actualmente, a mis 22 años, sigo explorando este mundo con la misma pasión, impulsado por la constante evolución de la tecnología y su interminable curva de aprendizaje.

Cuento con una carrera en Cómputo Forense y Ciberseguridad, actualmente curso una maestría en Inteligencia Artificial y Ciberseguridad.

Certificaciones: CWEE | CAPE |CPTS | EWPTX | CRTO | eMAPT | OSCP | OSCP+ |CEH V13 | EJPT| HTB prolabs Hades – Cybernetics – Zephyr – APTlabs | MDK

Creator: Cryptocurrency Village

Software supply chain attacks are no longer a distant threat — they are happening at scale and extremely dangerous from a crypto exchange’s POV.

As build pipelines grow more complex and dependencies multiply across npm, PyPI, SBOM, and internal registries, a single scanning layer is no longer enough to detect malicious code.

This talk focuses on an AI journey — from deploying a single AI agent to gate code merges, to architecting a full multi-agent system hardened through structured simulated exercises — along the way catching real-world attacks, including the coordinated compromise of a highly popular npm package spanning over 2 billion weekly downloads.

People:
    SpeakerBio:  Hanley Shun, OKX

Hanley has been working in information security for 10 years, with a background in penetration testing and vulnerability management. Now at OKX, channeling that offensive mindset into building secure CI/CD pipelines with AI, every day keeping customers’ funds safe.

SpeakerBio:  Cong Zhang, OKX

Cong has spent 5 years building SAST from the inside out — engines, rules, and the thinking behind them. From AST-level detection logic to full scanning pipelines, he knows where static analysis hits its limits. Now at OKX, he’s pushing past them with AI, teaching tools to reason about supply chain risks and logic flaws that pattern matching will never catch.

SpeakerBio:  ⁨Oscar Skjerven, OKX
No BIO available

Creator: AppSec Village

Let’s talk about some critical infrastructure that millions of people use every day: Public transportation. In this talk, I’ll present some findings that I discovered in the public transportation ecosystem of one of the largest cities in Argentina, impacting more than 1.5 million people daily. Reading code, chaining vulnerabilities, weak access controls, and flawed internal designs, I got full access to core mobility systems, from buses to taxis, including DVRs, transport cards, user data, real-time tracking and administrative panels. We’ll walk through the technical exploitation path, the real-world impact and the lessons learned.

People:
    SpeakerBio:  Ignacio Navarro

Ignacio Navarro, an Ethical Hacker and Security Researcher from Cordoba, Argentina. With around 6 years in the cybersecurity game, he’s currently working as an Application Security. Their interests include code analysis, web application security, and cloud security. Speaker at DEFCON, H2HC, Troopers, LeHACK, NorthSec, TyphoonCon, Security Fest, 8.8, among others. @Ignavarro1

Creator: Nix Vegas Community

How I got started with NixOS and how it’s going. I’ll cover my homelab migration from ubuntu & docker to NixOS & Incus w/ NixOS containers. I’ll also go over how a similar process is going at my $job, discussing packaging proprietary software into modules. I’ll cover declarative hardware and software in both domains from an SRE point of view.

People:
    SpeakerBio:  Jared

I’m an SRE based in $location working for $job.

Creator: Voting Village

Public trust in U.S. elections has been shaken by allegations that the 2020 presidential election was “stolen” or “rigged.” While many specific claims about that election are technically incoherent or demonstrably false, it is true that the way elections are administered in the U.S.–including eligibility determinations, voting equipment, procedures, and audits–leaves room for rational doubt about whether the reported outcomes of U.S. elections are correct. Given the known vulnerabilities in voting systems, the shortcomings of election procedures in some jurisdictions, and examples of large errors in vote counts and insider malfeasance, to blithely and automatically accept reported results is naive. To deserve public trust, elections should be “evidence-based”: conducted in a way that generates convincing public evidence that the reported outcome is correct. “Trust us” is not evidence. Evidence-based elections generally require demonstrably accurate eligibility determinations; recording votes primarily on hand-marked paper ballots; demonstrably secure chain of custody of voted ballots; a rigorous canvass to ensure that every validly cast ballot (and no others) was included; tabulation of the votes by hand or by machine; a compliance audit to ensure that the paper trail is trustworthy; and a properly designed, publicly confirmable audit of the reported results using the trustworthy paper trail. Conducting evidence-based elections requires avoiding the use of technology for purposes where outcome-altering malfunction or malfeasance could escape detection and correction, for instance, using electronic technology to record or transmit votes (e.g., paperless voting systems, touchscreen ballot-marking devices, cellular networks, or the Internet). Risk-limiting audits (RLAs) play an important role in evidence-based elections, providing affirmative evidence that the reported outcome is correct–or, with high probability, correcting the outcome if it is wrong. RLAs require a trustworthy paper trail. They cannot compensate for poor election administration, such as failing to keep track of voted ballots or using vulnerable technology to record or transmit votes.

People:
    SpeakerBio:  Philip Stark

Philip B. Stark is Distinguished Professor of the Graduate School at the University of California, Berkeley, where he has served as department chair and associate dean. In 2007 he invented “risk-limiting audits” (“RLAs”), endorsed by the National Academies of Science, Engineering, and Medicine and the American Statistical Association, among others, and required or authorized by law in about 15 states. He designed and helped conduct the first dozen pilot RLAs, helped draft RLA legislation for several states, and has published open-source software to support RLAs. In 2012, he and David Wagner introduced “evidence-based elections,” a paradigm for conducting demonstrably trustworthy elections. Stark has served on the Board of Advisors of the US Election Assistance Commission and its cybersecurity subcommittee, the Board of Directors of Verified Voting Foundation and the Election Integrity Foundation, and on the California Post Election Audit Standards Working Group. He has worked with the Secretaries of State of California, Colorado, and New Hampshire and numerous local election officials. He has testified about election integrity in state and federal courts and to legislators. He received the IEEE Cybersecurity Award for Practice, the UC Berkeley Chancellor’s Award for Research in the Public Interest, and the John Gideon Award for Election Integrity. He is a fellow of the American Academy of Arts and Sciences, the American Statistical Association, and the Institute of Physics.

Creator: Red Team Village

Modern endpoint defenses increasingly fingerprint .NET tools not only by strings and imports, but by structure: control flow, call graphs, and overall execution “workflows.” Traditional IL obfuscation can hide obvious indicators, yet often produces highly regular flattened patterns that are easy to flag on their own.

This talk presents a Roslyn-based, source-to-source obfuscator for C# that rewrites an entire project before compilation. Using the official compiler APIs, it transforms the code’s syntax and semantic structure, reshapes control/data flow, and then recompiles the result – while keeping the program’s observable behavior intact. The output tends to look complex but “natural,” diverging both from the original tool and from the telltale patterns created by many IL-level obfuscators.

I will walk through the pipeline, show how surprisingly little code is needed to build these transformations on Roslyn, and discuss what this means for EDR, sandboxing, and ML detections that rely on graph and workflow analysis of .NET binaries.

So, anyway, on moment of submitting original Seatbelt tool (obfuscated by source) have 10/73 security vendors flagged this file as malicious (can be better) https://www.virustotal.com/gui/file/ec2dcecb927874d41b051633135c8a594abb40d03f6836a548b40963bf330c86/detection/f-ec2dcecb927874d41b051633135c8a594abb40d03f6836a548b40963bf330c86-1764839678

My project https://github.com/gam4er/Loaders differs from more well known https://github.com/sadreck/Codecepticon by implementing semantic structure obfuscation.

My point of view (sa a SOC analyst, blue teamer) on “how to hide mytool from AV” problem is: if you have a code – OBFUSCATE YOUR CODE. Stop modifying binaries

People:
    SpeakerBio:  Alexander Rodchenko

Rodchenko Alexander is a Senior SOC Analyst at the SOC Security Research Group at Kaspersky. He began his career at OJSC Rosneft, focusing on industrial safety, troubleshooting, and audits. Currently, he investigates industry events and trends with the primary goal of integrating these insights into monitoring and threat hunting practices. Leveraging his extensive expertise, Alexander advises customers and threat detection/hunting teams on the optimal response to emerging threats and trends. In addition to speaking at Positive Hack Days (twice) and BSides Zurich 2023, he has also been a speaker at CodeBlue 2024 and BlackHat MEA 2024.

SpeakerBio:  Ashley Hiram Muñoz, Kaspersky – Incident Response Specialist

I currently work as an Incident Response Specialist on Kaspersky’s Global Emergency Response Team (GERT). I live in Mexico and have over seven years of experience in Incident Response, Digital Forensics, Malware Analysis, and Reverse Engineering. Before joining DFIR, I worked for two years as a Penetration Tester.

I have collaborated on various Threat Hunting and Threat Intelligence projects.

Additionally, I have been a speaker at international events such as DEFCON (La Villa Hacker), BSides, Ekoparty, 8.8, HackGDL, BugCON, Pwnterrey, and others. I currently teach the Digital Forensics, Malware Analysis, and Incident Response modules in an information security diploma program at UNAM (Universidad Nacional Autónoma de México).

Certifications: GREM, GCFA, GCFR, eCTHP, CHFI.

—

Actualmente me desempeño como Incident Response Specialist en el Global Emergency Response Team (GERT) de Kaspersky, cuento con +6 años de experiencia realizando Respuesta a Incidentes, Análisis Forense Digital, Análisis de Malware y Reversing; previo a dedicarme a DFIR laboré 2 años como Penetration Tester.

He colaborado en distintos proyectos de Threat Hunting y Threat Intelligence.

Adicionalmente, he sido ponente en eventos internacionales como DEFCON (La Villa Hacker), BSides, Ekoparty, 8.8, BugCON, etc.

Actualmente soy profesor de los módulos de Análisis Forense, Análisis de Malware y Respuesta a Incidentes en un diplomado de seguridad de la información de la UNAM.

Certificaciones: GREM, GCFA, eCTHP, CHFI.

SpeakerBio:  Eduardo Chavarro Ovalle, DFIR Group Manager at Kaspersky – GERT

Eduardo Chavarro Ovalle, DFIR Group Manager for Americas, member of Kaspersky GERT Team. Student of DBA in ML and AI and MSc in Cybersecurity with more than 20 years of experience in cybersecurity, DFIR, eDiscovery, and threat analysis. GCIH | GRID | GCFA | CISM | CHFI | CPTE | SFCP | ITIL.

Creator: Red Team Village

What if a rogue access point could evade a Wireless Intrusion Prevention System (WIPS) while operating in plain sight?

In this hands-on tactic, participants will learn how EvilEssid, a wireless evasion technique, exploits weaknesses in how WIPS solutions identify, classify, and trust wireless networks. Rather than focusing solely on theory, attendees will actively build, configure, and operate EvilEssid in a controlled lab environment. Through a series of guided exercises, participants will gain practical experience deploying rogue access points, understanding WIPS detection logic, and applying wireless identity manipulation techniques to evade common defensive controls. The workshop will demonstrate how attackers can establish malicious wireless infrastructure that remains undetected while creating opportunities for credential harvesting, traffic interception, and adversary-in-the-middle operations. Attendees will use the EvilEssid tool to reproduce the attack chain step-by-step. Each exercise is designed to provide hands-on exposure to offensive wireless tradecraft while encouraging participants to experiment. Beyond the offensive techniques, the workshop will also explore defensive implications, helping participants understand the limitations of current WIPS technologies and identify practical approaches to improve wireless threat detection. By the end of the session, attendees will have deployed a rouge access point using EvilEssid themselves, validated successful WIPS evasion in a lab environment, and gained a deeper understanding of both the offensive and defensive aspects of advanced wireless operations.

Fundamentals • ESSID
• BSSID
• Deauthentication attacks • Evil Twin attack Understanding WIPS and Its Detection Models • What is a WIPS • How it works and detection techniques • Evil Twin attack vs WIPS Identifying the Weakness Advancing the Evil Twin Technique • Human targets • Crafting a WIPS-evasive Evil Twin EvilEssid tool: Design and Implementation • Overview • Key concepts behind the tool • Features and usage Conclusions and Countermeasures

People:
    SpeakerBio:  Miguel Fernandez

Colombian cybersecurity researcher based in China with over 15 years of experience in offensive security, penetration testing, and applied security research.

His work focuses on developing and validating novel attack and defense techniques, challenging existing security assumptions across areas such as wireless security, social engineering, and real-world adversarial methodologies.

He has presented at international conferences including Codegate, Overdrive, CIS, Xcon x Hacking group ,WAIC, Hackprove world, Tencent security saloon and multiple PRC cyber security meetups.

His research is grounded in the scientific method, emphasizing experimentation, reproducibility, and practical validation.

Creator: Bug Bounty Village

For two decades, XSS was the king of data theft — exploiting trust boundaries in a victim’s browser to exfiltrate sensitive data and perform malicious actions. AI-powered interfaces have introduced new ways to deliver XSS to victims alongside an entirely new bug class that has emerged as the most impactful vulnerability pattern in modern AI applications. The parallels between XSS and AI data exfiltration are striking — both exploit trust boundaries to force a system into performing unintended actions on behalf of a victim but the attack surface in AI is broader, spanning web interfaces, mobile apps, wearables, and any client consuming agent output. The exfiltration channels are diverse: markdown image rendering, iframe and HTML tag injection, sandbox escapes, native platform connectors, network connectivity tools, javascript: URI schemes, link unfurling side channels, MCP server poisoning, and multi-step agent tool-abuse chains. In this talk, two bug hunters who’ve spent the past year exploiting these vulnerabilities across production applications break it down and walk through real attack chains, demonstrate how we bypass common mitigations and share the bypass techniques we’ve developed. We’ll cover how to threat model AI applications, identify viable attack paths, and select delivery mechanisms to weaponize end-to-end exploits. Beyond offense, we’ll discuss the mitigations we’ve encountered, what actually works, what doesn’t, and emerging trends in agentic AI and autonomous tool use mean for the next wave of vulnerabilities hunters should be watching for. Leave with: (1) a clear mental model of AI data exfiltration as a bug class and how it relates to familiar web primitives, (2) concrete attack techniques and bypass methods, (3) practical threat modeling for identifying exfil paths in AI agents and applications, and (4) visibility into upcoming trends and bug classes they should be hunting as AI systems gain more autonomy and connectivity.

People:
    SpeakerBio:  Ads Dawson, Staff AI Security Researcher, OWASP GenAI Security Project founder

Ads Dawson founded the OWASP GenAI Security Project and led it to flagship status — the fastest in OWASP history. As a Staff AI Security Researcher at Dreadnode, he specializes in exploiting ML and AI systems, harnessing AI for offensive cybersecurity through capability development and exploit development, and conducting red team operations against frontier models for government, military, and tier-1 labs. He is lead author of AIRTBench (arXiv), the first benchmark for autonomous AI red teaming in LLMs, and author of AI Native LLM Security (Packt, 2025).

A senior red team operator with BT6 (the frontier AI red team), ranked #2 in Canada on HackerOne (2025/2026), and selected for Meta’s MBBRC live hacking event, Ads is a HackerOne US South Ambassador, BugCrowd Hacker Advisory Board member, MITRE AI Working Group contributor, and leads the OWASP Toronto chapter. He spoke at Bug Bounty Village DC33 on the BT6 AI jailbreaking panel and is also presenting “Exfil Everything: A Year of Stealing Data from AI Agents” at Bug Bounty Village DC34 (schedule pending publication).

SpeakerBio:  Mike “TakSec” Takahashi, AI Red Team Researcher, Zenity

Mike Takahashi, aka TakSec, is an AI Red Team Researcher at Zenity; member of BT6; and Bug Bounty Hunter focused on breaking AI systems. He has submitted 400+ vulnerabilities across major bug bounty programs and top ranking on both Anthropic’s Safety Bug Bounty Program on HackerOne and Mozilla’s 0din GenAI Bug Bounty Program. His work targets prompt injection, data exfiltration, and AI agent security.

Creator: Red Team Village

Abstract—This presents an exhaustive empirical security analysis of a real-world private 5G network deployment, serving as a direct and explicit extension of our preceding research, ”Security Analysis of a Real-World Private 5G Network Deploy- ment through Pen Testing and Fuzzing.” While our initial study successfully identified control-plane weaknesses and performance degradation via protocol mutation, this continuation focuses on four critical, highly exploitable attack vectors. Specifically, we detail three severe data-layer attacks stemming from the unauthenticated internal network exposure of MongoDB (Port 27017), Elasticsearch (Port 9200), and Kibana (Port 5601). These exposures permitted the complete, unauthorized extraction of the network’s most sensitive cryptographic assets (including Permanent Keys and active session keys) and granted adversaries unauthorized manipulation of the security monitoring apparatus. Additionally, we introduce a newly discovered, critical Denial of Service (DoS) attack targeting the Access and Mobility Management Function (AMF) signaling interface on Port 11000, which leverages protocol fuzzing to induce memory corruption and permanently halt User Equipment (UE) registration. By synthesizing these empirical findings, this report provides a multi- layered analysis of the cascading failures that emerge when enterprise IT vulnerabilities intersect with mission-critical 5G telecommunications infrastructure.

People:
    SpeakerBio:  Aumkaareshwar DS

I am a Computer Science graduate student at California State Polytechnic University, Pomona, passionate about cybersecurity, network security, and ethical hacking. Currently, I work as a Research Assistant at PolySec Lab, where I focus on 5G network security, including authentication, subscriber identity protection, and threat mitigation. My research helps enhance mobile security and protect data from cyber threats.

Creator: Blacks In Cyber Village

Join us for an in-depth session exploring the critical security features embedded within the Armv8-M architecture. We’ll dive into TrustZone for Cortex-M, the PACBTI Extension, and the Debug Authentication/Unprivileged Debug Extension, providing clear explanations of their functionalities and the motivations behind their development. Discover how these advanced features contribute to robust embedded system security.

People:
    SpeakerBio:  Ian Harris, University of California, Irvine

Ian Harris is a Professor at the University of California, Irvine, specializing in computer science with a strong focus on embedded systems and security. They bring a rich background in hardware design verification and security, and their research encompasses secure hardware/software systems. Ian also teaches courses related to embedded systems and has presented at cybersecurity conferences.

SpeakerBio:  Jacob Gutierrez, Cybersecurity Professional / Community Speaker

Jacob Gutierrez is a recent Computer Science graduate with a keen interest in embedded systems and their security. They are passionate about contributing to the cybersecurity community and sharing their knowledge.

Creator: ICS Village

As data center numbers are increasing in the US and around the world, they are becoming a source of tension politically, environmentally, and economically, and are becoming high-value critical infrastructure. The threats to these data centers will rise soon as activists, cybercriminals, and nation-states target them. Unlike what has been seen recently with kinetic attacks on data centers in conflict zones, cyberattacks have a larger range and can affect a wider population.

We wanted to explore unconventional threats against data centers. Using open-source intelligence and our patented (US12267344B1) geostalking techniques, we were able to map out data centers in the United States and overlay that with the exposed infrastructure that would directly be used in the operation of the data center itself. This includes building automation and energy supply as examples.

Afterwards, we were left with the locations of 1,063 data centers that had 6,300 high-confidence industrial control systems exposed to the internet. This information went through a five-layer filtering process to result in these high-confidence exposed systems, based on the banner responses received from these devices. Of these 6,300 devices, 964 devices (15.3%) have a CVSS score of 9.0 or above, and 88.7% of the entire dataset is inherently vulnerable due to the use of protocols that were never designed to be exposed to the internet.

Armed with this information, an attacker could circumvent even the best IT security with exposed systems that might directly or indirectly affect data center operations. In today’s geopolitical climate, exposed systems in near proximity to data centers are at higher risk of cyberattacks. Due to diverse applications across personal, corporate, or even government use, the repercussions of a data center outage will have effects far wider than just the data center itself.

People:
    SpeakerBio:  Stephen Hilt

Stephen J. Hilt is a seasoned threat researcher specializing in industrial control systems (ICS) security, cybercrime investigations, and advanced persistent threats. With extensive experience in uncovering complex attacks targeting critical infrastructure, he has contributed to numerous high-impact reports and global threat intelligence initiatives. Stephen’s work bridges deep technical expertise with strategic analysis, enabling organizations to better understand and defend against emerging cyber threats. His research has been presented at leading security conferences and published in widely respected industry reports, reflecting his commitment to advancing the security of interconnected systems worldwide.

SpeakerBio:  Numaan Huq
No BIO available

Creator: AppSec Village

Shared threat frameworks have enabled consistent communication and coordination across vulnerability researchers, PSIRTs, vendors, and defenders. However, as attackers increasingly operate at the application layer (abusing software supply chains, runtime behavior, and trusted cloud workflows) defenders face challenges applying existing models consistently. This session presents lessons from the Application Attack Matrix, a community-driven research effort involving contributors from Mandiant (Google Cloud), Microsoft, AWS, Meta, Oligo Security, and others, and uses the ByBit / Safe{Wallet} incident to illustrate why application-layer technique modeling is increasingly important.

People:
    SpeakerBio:  J Fridley

J. Fridley is a Senior Solutions Engineer at Oligo Security, helping teams secure modern cloud environments with deep runtime visibility. He previously led and supported Solutions Engineering teams at Snyk and has a background as a software engineer and technical project lead. J also brings nearly two decades of military service with the U.S. Navy and Army National Guard.

Creator: Red Team Village

Chrome controls over 65% of worldwide browser usage, with 100,000+ extensions on the Chrome Web Store. Enterprises rely on Chrome’s built-in policies, such as extension allowlisting, WebStore integrity checks, and forced installed extensions to keep their endpoints clean. Those controls are broken. Extension Hollowing is a new technique that completely bypasses Chrome’s integrity verification of WebStore extensions. By hollowing out a legitimate allowlisted extension and replacing or backdooring its internals, an attacker can silently deploy extension arbitrary code inside the browser, defeating allowlisting, bypassing signature checks, and inheriting the full permission set of the original extension (and extending it if desired). This talk will show how hollowed extensions can function as a fully capable C2 via a new Chrome extension mythic agent which will be released after this talk called Hades. Attendees will get to sit down and play with this c2 framework as well as embedding it into trusted extensions via extension hollowing. We will then walk through the specific Chrome enterprise policies designed to prevent this, and how to bypass them. The technique extends beyond Chrome to other Chromium-based browsers including Edge. TLDR; This talk will demonstrate live exploitation, discuss what EDRs are (and aren’t) keying on, cover which mitigation strategies actually work, and why most deployed configurations don’t. In addition attendees will get to sit down and actually embed a custom Chrome extension mythic agent inside of trusted extensions, bypassing active policies and showing how to dynamically load and run code in the latest versions of Chrome.

People:
    SpeakerBio:  Gordon Long

Gordon Long is the President and CEO of LegioX Cyber Technologies and is a Senior Offensive Security Engineer at Zoom. He also teaches as an Adjunct Professor at George Mason University in the Digital Forensics Master’s Program. His favorite areas of cybersecurity include EDR evasion and cyber deception. His X handle is @ethicalhax and github is under AsaurusRex.

Creator: Red Team Village

Falco is one of the most widely deployed cloud-native runtime security tools, relied upon to detect suspicious behavior across containers and Kubernetes environments. But what happens when attackers understand the rules better than defenders? In this hands-on tactic, attendees will compromise a Kubernetes environment while interacting directly with Falco detections in real time. Participants will execute common attacker actions including secret harvesting, privilege escalation attempts, reverse shell execution, and lateral movement, then iteratively adapt their tradecraft to evade or blend around runtime detections. Along the way, attendees will learn how Falco rules work, where syscall-based detection succeeds, where it struggles, and how attackers abuse assumptions in default rule sets. The session emphasizes practical detection engineering lessons for both offensive and defensive practitioners.

People:
    SpeakerBio:  Michael Reimsbach, Product Security Specialist at SAP

Michael is a Product Security Specialist at SAP, working with the SAP Cloud Infrastructure security team. His focus areas include vulnerability management, secrets management, and building secure internal services.

He obtained multiple industry certifications such as OSCP, GCPN, and CISSP.

A healthy dose of paranoia led him to explore OSINT and the surprising power of publicly available information.

Beyond his day-to-day work, Michael is an active member of the cybersecurity community and helps organize BSides Luxembourg.

Creator: Adversary Village

Offensive agents are easy to demo, but much harder to make useful across real adversary emulation work. Key challenges lie in making agents reusable and reliable across different assessment types, tech stacks and doing so, as models, tools, and environments change.

This talk shares lessons from designing, deploying, and refining a suite of modular offensive agents used in real client engagements. We will cover what made the agents reusable, where they broke against real-world complexity, and how we evaluated their effectiveness over time as prompts degraded, models drifted, and underlying tools evolved. The session is framework-agnostic and vendor-neutral.

Attendees will leave with a practical model for building composable offensive agents that remain useful and reliable across attack types.

People:
    SpeakerBio:  Dominika Pietrzak, Offensive AI Engineer

Building AI-led tooling to scale offensive security operations.

SpeakerBio:  Ibai Castells, Senior Red Team Hive Member

Senior Red Teamer and Red Team Capability Lead at CovertSwarm. Passionate about Windows, AD, malware dev and offensive tool engineering. Currently exploring how AI can be integrated in Red Team workflows. I drop my latest research, malware dev, and how-tos on my blog and GitHub.

Creator: Policy @ DEF CON

People:
    SpeakerBio:  Katherine Pratt, Microsoft

Katherine Pratt is a senior security engineer and operator on the AI Red Team at Microsoft. She received her B.S. in aerospace engineering from MIT in 2008. Following graduation, she served four years in the United States Air Force, spending most of her time as an operational flight test engineer on the F-35 Joint Strike Fighter. She received a PhD in Electrical and Computer Engineering from the University of Washington in 2019, where she studied the privacy, ethics, and policy of neural data. Her professional work experience includes Blue Origin, the ACLU of Washington, and a fellowship through TechCongress working on technology policy in the US House of Representatives. She also competes in triathlons and is the co-founder of the 501(c)3 Minority Veterans of America.

SpeakerBio:  Jeff Rothblum, Founder at Plaintext Strategies

Jeff Rothblum, founder of Plaintext Strategies, has worked at the intersection of technology and policy for two decades. He has led government affairs at an AI startup, served as a cyber threat intelligence analyst, led cybersecurity policy the Senate Homeland Security and Governmental Affairs Committee, served as a Director of Cyber Policy and Plans at the White House Office of the National Cyber Director, and co-owned an artisan blacksmithing company.

SpeakerBio:  Maurice Turner

Maurice Turner is a recognized public interest technologist and cybersecurity expert on the Public Policy team at TikTok. As a Technical Policy Lead, Turner plays a key role as a liaison collaborating between public policy and cross-functional teams to ensure understanding of cutting-edge technologies by non-technical audiences. He also advises the broader Americas public policy team on a wide range of technology, strategy, and policy issues.

SpeakerBio:  Ayan Islam

Ayan Islam is an Adjunct Lecturer at American University School of Public Affairs’ Cybersecurity Policy and Management program and cyber risk specialist at a Fortune 500 FinTech company. As former White House Director of Cyber Workforce, she served at the Office of National Cyber Director (ONCD) supporting the development and implementation of the National Cyber Workforce and Education Strategy. Previously, she was the Associate Policy Director for R Street’s Cybersecurity and Emerging Threats team and contributed to special initiatives at the Department of Homeland Security (DHS) Cybersecurity and Infrastructure Security Agency (CISA).

SpeakerBio:  Beau Woods
No BIO available

Creator: AppSec Village

AI made finding and exploiting bugs almost free. In one month, Anthropic’s Project Glasswing surfaced over 10,000 high and critical bugs and patched fewer than 100. The maintainers on the receiving end are asking Anthropic to slow down, because they can’t fix the bugs fast enough.

If one model can bury the world’s best-funded security teams, what happens to the 1-to-5 person team with little security budget and a backlog now growing at machine speed?

This talk shows you how to build SPAR (Scan, Prioritize, Auto-fix, Rewrite): a remediation pipeline that chains call-graph reachability, EPSS percentiles, and SSVC decision logic into triage a small team can run, then routes the ones that matter through AI patch generation behind a regression-aware verification harness.

People:
    SpeakerBio:  Mohan Kumar

Mohan is a security leader with over a decade of experience in security architecture, engineering, and operations. He has a strong interest in developing robust security programs and a proven track record of creating proactive security roadmaps and strategies aligned with business objectives. He constantly seeks ways to elevate security processes and culture to the next level.

SpeakerBio:  Naveen

Naveen is a Security Researcher with over 9 years of expertise specializing in AI, application, and cloud security. He invented 5 patents in Agentic Security space.

He possesses extensive knowledge in all aspects of product security, including threat modeling, DevSecOps, API security, and penetration testing. He is passionate about integrating security into the SDLC from design to deployment, ensuring the early detection and mitigation of vulnerabilities.

Creator: ICS Village

What does the global attack surface of operational technology actually look like at internet scale? We built IRONMAP, a purpose-built OT/ICS intelligence platform, to find out — and the answer is both larger and more disturbing than we expected.

Over the course of this ongoing research project, IRONMAP has catalogued over 5.5 million ICS/OT-facing assets across the public internet, with more than 2.25 million flagged as high-risk. Using deep protocol fingerprinting across all major industrial protocols — EtherNet/IP (CIP), Modbus TCP, Siemens S7, DNP3, IEC 60870-5-104, OPC-UA, BACnet/IP, Omron FINS, GE SRTP, Tridium Fox, and more — IRONMAP goes well beyond port scanning to perform authenticated protocol enumeration, live register reads, and tag harvesting using PLCDISCO, our multi-protocol OT scanner.

This talk presents a ground-level statistical portrait of the exposed OT internet: which protocols dominate, which sectors are most exposed, how vendor market share looks through the lens of deep insight and where in the world the highest concentrations of exposed critical infrastructure live (spoiler: it is not all China). We will walk through what over 242,000 EtherNet/IP devices look like when you enumerate their CIP identity objects, what ~500,000 Modbus devices expose in their holding registers, and what the live tag names of real PLCs tell you about what processes they are running.

We then turn to a specific and underappreciated issue: Automatic Tank Gauges (ATGs). IRONMAP found 149 confirmed ATG systems directly exposed to the internet, the majority of them Veeder Root TLS-350 and TLS-450 units — the dominant ATG platform at commercial fueling facilities across North America. These systems, reachable via the Guardian ASP protocol on TCP/10001, require no authentication on older firmware and respond to a simple serial-style command set with:

• Current fuel volume per tank, in gallons
• Product type (Unleaded, Premium, Diesel, Jet-A, Heating Oil)
• Live alarm states (high water, leak detection, low fuel, overfill)
• Delivery event history and timestamps
• Up to 8 tank probes per unit

The implications are significant. Exposed ATGs reveal not just that a facility has fuel storage, but how much, what kind, and when deliveries occur — operational patterns that are directly relevant to physical security and supply chain intelligence. IRONMAP discovered ATGs at locations that include commercial truck stops, bulk fuel terminals, and sites with product profiles consistent with aviation or military use. We will demonstrate a live walk-through of what an unauthenticated session reveals, discuss the responsible disclosure posture we have taken, and present mitigation guidance for asset owners.

Attendees will leave with a realistic, data-grounded view of the exposed OT landscape — not a cherry-picked set of scary screenshots, but statistically representative findings from a 5.5-million-asset dataset — plus actionable context on the ATG exposure class and how to find and fix it.

People:
    SpeakerBio:  Matt Caldwell, Tophat Security

Matt Caldwell is the founder of Tophat Security, cyber security firm specializing in innovative OT/ICS software and tools, red team operations, and critical infrastructure research. With over a decade of experience in industrial control system security, Matt has assessed environments spanning oil and gas, electric utilities, water treatment, manufacturing, and federal government. He built IRONMAP as a research platform to continuously map the internet-exposed OT attack surface at scale, applying it to build threat intelligence, support disclosure efforts, and develop data-driven visibility into the global ICS exposure problem. Matt holds relevant certifications in cyber security and penetration testing and has presented research at global OT/ICS security conferences. He is based in Athens, Georgia, and speaks regularly with asset owners, government stakeholders, and research organizations on OT exposure topics.

Creator: Physical Security Village

Physical red teaming remains one of the most consequential and under-discussed disciplines in offensive security. While network intrusions dominate the conversation, a determined adversary at the perimeter can bypass millions of dollars in cyber defenses with a lanyard, a clipboard, and a confident smile. This talk distills lessons learned from real-world physical red team engagements that achieved their objectives, walking through the tactics, techniques, and tooling that consistently produced results across diverse target environments.

Topics include reconnaissance and target profiling, pretext development and social engineering at entry points, covert entry tooling (bypass devices, RFID cloning, lock and latch attacks), implant deployment for persistence, and methods for chaining physical access into network footholds. Emphasis is placed on what actually worked, not theoretical attack trees, alongside the operational considerations, failure modes, and decision points that separate a successful op from a burned one.

People:
    SpeakerBio:  Michael “v3ga” Aguilar, Principal Consultant at Sophos Red Team

Michael Aguilar (v3ga) is a Principal Consultant on the Sophos Red Team, paid to think like the people his clients are afraid of. His work lives at the intersection of offensive security, vulnerability research, and low-level systems programming, Windows internals, reverse engineering, and the kind of dusty attack surfaces that nobody has looked at since the protocol was ratified. Sometimes, he’s lucky enough to perform full Physical Red teams against his client targets. His method is unglamorous and effective: Analyze, Formulate, Target, Attack (covertly).

Creator: AI Village
This is a sit down discussion in a more casual conversational format: Companies are increasingly deploying coding agents to triage bug reports, resolve support tickets, and open pull requests. These agents have direct access to codebases, CI/CD pipelines, internal tooling, and, often, the internet. These are exactly the conditions that make indirect prompt injection devastating.

This talk presents end-to-end vulnerability chains we have discovered and disclosed in real coding agents. We’ll demo how adversary-controlled content (e.g., a bug report from a user) can hijack an agent’s goal and lead to outcomes like source code exfiltration and malicious pull requests.

As coding agents become increasingly autonomous and interact with the outside world more frequently, these attacks will only grow more potent. We close with recommendations for how companies and codegen providers can defend against them.

People:
    SpeakerBio:  Matt Galligan
Matt has spent his career on both sides of the fence: building deceptive cyber systems and the security automation behind one of the DOD’s largest software factories, then joining CISA’s Rapid Action Force — where the job was, more or less, “hack the entire federal government, daily,” between vuln research and no-notice red team engagements. After a stint in big tech, he’s now at Corridor focusing his efforts on building a rare thing in the industry: a security product that security teams actually love.
SpeakerBio:  Jack Cable, CEO and Co-Founder at Corridor

Jack Cable is the CEO and Co-Founder of Corridor, the security layer for AI coding. Prior to that, Jack served as a Senior Technical Advisor at the Cybersecurity and Infrastructure Security Agency (CISA), where he helped lead the agency’s Secure by Design initiative. Before CISA, Jack worked as a TechCongress Fellow for the Senate Homeland Security and Governmental Affairs Committee, advising Chairman Gary Peters on cybersecurity policy, including open source software security. He previously worked as a Security Architect at Krebs Stamos Group. Jack is a top bug bounty hacker, having identified over 350 vulnerabilities in hundreds of companies. After placing first in the Hack the Air Force bug bounty challenge, he began working at the Pentagon’s Defense Digital Service. Jack studied computer science at Stanford University and has published academic research on election security, ransomware, and cloud security.

Creator: Social Engineering Community Village

What happens when you social engineer an AI agent that was trained to be helpful over the phone? Can you get it to reveal its system prompt out loud? Will it disclose information about other callers? How far can you push it before its guardrails kick in? AI voice agents sit behind telephony layers like speech-to-text, text-to-speech, and call routing that introduce new attack surfaces and opportunities. They’re replacing human operators everywhere: answering phones at doctors’ offices, handling IT help desks, triaging customer support, and booking appointments. They sound human, but underneath, they’re the same LLMs we’ve been prompt injecting. I built an open-source tool that tackles this by placing real phone calls to voice AI agents, speaking attack scenarios using text-to-speech, capturing responses via speech recognition, and analyzing transcripts for signs of successful exploitation. It maps 20 attack scenarios across five categories from the OWASP Top 10 for LLM Applications: prompt injection, sensitive information disclosure, system prompt leakage, excessive agency, and misinformation. Detection uses pattern matching and an LLM judge to catch both obvious and subtle failures.

People:
    SpeakerBio:  Willie Zhang, Offensive Security Consultant

Willie Zhang is an Offensive Security Consultant with experience protecting companies by thinking like an attacker. What started as a $1 online course on ethical hacking in college turned into a career built on finding the gaps in systems that aren’t supposed to have any. Willie has a growing passion for understanding and attacking AI systems, building on a foundation of testing everything from corporate networks to the humans that run them. When he’s not learning something new, Willie is in a League of Legends lobby, because apparently cybersecurity isn’t chaotic enough.

Creator: Data Duplication Village

Traditional software-based security cannot prevent data destruction or unauthorized cloning when authorized credentials are compromised. This presentation introduces a hardware-level physical intervention approach with a zero-OS FPGA architecture to overcome vulnerabilities at the L3/L4 layer. This invisible Layer-2 transparent bridge, lacking IP or MAC addresses, performs sub-nanosecond-level gate-level Deep Packet Inspection (DPI) on the pipeline. When high-risk storage commands (deletion, unauthorized data transfer, etc.) are detected at the silicon level, the data flow is hardware-intercepted and held in a “Catch-and-Release” buffer until a physical human confirmation is received.

People:
    SpeakerBio:  Mehmet Önder Key, Cyber Security Consultant

Önder Key is a cybersecurity consultant specializing in critical infrastructure security, zero-day vulnerability analysis, and offensive security. He has advised organizations in high-security sectors such as defense, aerospace, and finance, with hands-on experience in both red teaming and strategic security engineering. His work has been featured across numerous countries and platforms, contributing to the discovery of systemic vulnerabilities. Currently, he provides consultancy to TurkNet and continues to advance the global offensive security ecosystem by challenging traditional approaches to cybersecurity.

SpeakerBio:  Temel Demir

Temel Demir is a cybersecurity researcher and hardware architect specializing in hardware defense, offensive security, and the protection of critical infrastructure. With a core focus on embedded system vulnerabilities and Layer-1/Layer-2 network manipulation, his research involves developing deterministic, hardware-enforced frameworks that bypass traditional software-defined security flaws. Having previously shared security research on global stages, his work bridges the gap between sophisticated threat actor methodologies and immutable physical security barriers.

Creator: Physical Security Village

You know lockpicking. You know lock bypass. You might know evasion tactics for alarms and surveillance systems. But what traces get left behind if an attacker does these? This talk will give an introduction to the forensics of covert entry techniques – what marks get left behind, how to tell whether someone entered, when it happened, what tactics and tools were used, and what skill level the attacker had. Taught by two founding members of the Physical Security Village, this talk will also show the interesting and funny results when certain entry techniques are applied thousands of times on one door, which will tune your eye for spotting when it’s only happened once.

People:
    SpeakerBio:  Bill Graydon, GGR Security

Bill Graydon is a Principal at GGR Security, where he designs and evaluates security systems and building construction for a wide range of industries – including testing them on physical pen tests. Through this, he hacks everything from locks and alarms to critical infrastructure; this has given him some very fine-tuned skills for breaking stuff. He’s passionate about advancing the security field through research, teaching numerous courses, giving talks, and running DEF CON’s Lock Bypass Village. He’s received various degrees in computer engineering, security, and forensics and comes from a broad background of work experience in cyber security, anti-money laundering, and infectious disease detection.

SpeakerBio:  Bobby Graydon

Robert is an avid researcher of lock manipulation, picking, bypass, and other vulnerabilities to discover and evaluate possible flaws and methods of attack. He has a passion not only for hacking, but teaching as well, being one of the founders of Physical Security Village, and enjoys speaking at various physec engagements. He has well-honed skills, as well as a thorough understanding of the mechanics and function of many types of high security locks, allowing him to effectively search for and test methods of cracking high security systems.

Creator: OWASP Foundation

We are moving beyond prompt engineering into the era of Agentic Warfare. This session pulls back the curtain on a clandestine bridge between the OWASP community and the underground. We aren’t here to lecture on “best practices.” We are here to hand you the blueprints for the Forge.

The Agenda:

The Artisan’s Edge: Weaponizing LLMs as high-bandwidth force multipliers for surgical, manual exploit crafting. Agentic Persistence: Engineering self-correcting, goal-oriented malware that adapts to environment constraints in real-time. Ghost Pipelines: Architecting air-gapped, invisible infrastructures for weapon synthesis where the trail ends before it begins. Monetizing the Machine: Scaling high-impact, client-side payloads for maximum “fun and profit.”

OWASP has chapters in every major city—think of them as your local cells. We are opening the laboratory doors to the elite. If you have the craft, we have the infrastructure.

Come claim your seat in the Forge.

People:
    SpeakerBio:  Jon McCoy, Security Architect, OWASP

Software security architect, Jon McCoy brings over 20 years of experience in software development and cybersecurity to the forefront. With a strong foundation in .NET development, Jon transitioned into security, driven by a passion for proactive defense strategies and secure coding practices.

A dedicated contributor to the OWASP community, Jon has shared his expertise at numerous industry events, including OWASP Global AppSec. His recent presentation on “Lessons Learned from Past Security Breaches” highlighted critical takeaways for strengthening AppSec efforts before and after incidents.

SpeakerBio:  Andra Lezza, Principal Application Security Specialist at Sage / OWASP Leader

Andra is a Principal Application Security Specialist at Sage, with over seven years of experience in the field of application security. She is responsible for implementing DevSecOps practices, conducting security assessments, and developing secure coding guidelines for software engineering and AI/ML teams. She has a strong background in software development and project management, as well as a master’s degree in information and computer sciences. She has been co-leading the OWASP London Chapter since 2019, where she organises and delivers events and workshops on various security topics. She is passionate about educating and empowering developers and stakeholders to build and deliver secure software and best practices in a fast-paced, results-driven environment.

Creator: Crypto & Privacy Village

Despite securing the connections of over a billion people, Russian and Chinese state cryptography often remains unfamiliar to western practitioners. We survey the GOST, ShangMi, and ZUC standards, their respective histories, designs, and the current academic understandings of their security. We additionally cover implementation conveniences, such as the use of AES-NI instructions with SM4 as a method of side channel hardening, and algorithmic peculiarities, such as the hidden substructure of the Kuznyechik and Streebog S-box. We conclude with considerations about why specific algorithms were standardized, and how they reflect the priorities of the respective nations that standardized them.

People:
    SpeakerBio:  1nfocalypse

1nfocalypse is a GCC developer and cryptography enthusiast, with a particular focus on applied cryptography and implementation security. He has contributed to libstdc++-v3 and GCC’s OpenMP implementation, authored cryptographic CTF challenges for C2 Society and the “?Cube” CTF, and has conducted security research leading to CVEs involving cryptographic aspects of Apache Druid and mbedTLS.

Creator: Physical Security Village

As physical security threats increase in sophistication and frequency, organizations face a critical gap in their physical security, the lack of a structured framework for physical security assessments and tests. FORTRESS was designed to fill this need by providing a categorized model of physical Tactics, Techniques, and Procedures (TTPs), each aligned to industry-standard compliance controls such as NIST, HIPAA, FedRAMP, ISO 27001, and PCI DSS. Designed for both red and blue teams, FORTRESS enables coordinated, repeatable physical security testing by consolidating adversarial behaviors into mapped TTPs. Each TTP is supported by detailed guidance for the execution, validation, and reporting of physical security gaps making assessments more consistent across multiple teams, business units, and future engagements. This paper presents the overall breakdown of FORTRESS, highlighting its application, structure, and flow. I believe this framework fills that critical void in operational and physical security and can be viewed as the foundation for enhancing and maturing physical security risk programs with real threat models as the base.

Links:
    Website – https://fortressframework.com/

People:
    SpeakerBio:  Brad “Sno0ose” Ammerman

Brad Ammerman, a leading figure in security testing, currently serves as the Senior Director at Prescient Security. His background includes influential roles at companies like Foresite, Optiv, Lockheed Martin, DIA, DoD, and Supreme Court of Nevada, where he developed his expertise in offensive security and team management. A skilled hacker himself, Brad is also a recognized speaker, educator, mentor, and disabled veteran, dedicated to teaching and protecting others.

Creator: Biohacking Village

Medical devices are becoming increasingly connected—and that includes devices responsible for keeping people alive. In this talk, we share our journey as four security students taking on the challenge of analyzing an insulin pump as relative newcomers to hacking medical devices. Motivated by curiosity, concern for patient safety, and personal stakes, we set out to explore how an attacker might approach such a system using only public information, basic wireless knowledge, and persistence.

Rather than presenting ourselves as experts, we focus on the learning process: how we approached an unfamiliar, safety critical system, how we performed threat modeling when the “failure mode” is a human body, and how we handled the many moments where everything stopped making sense. We’ll walk through what worked, what didn’t, and how critical thinking helped us move forward when we hit a wall.

By reflecting on where we started, where we are today, and what remains unexplored, this talk highlights the value of a beginner’s mindset when analyzing real world systems like medical devices. Our goal is not to sensationalize risk, but to show how accessible security research, done responsibly, can contribute to better understanding and safer technology.

People:
    SpeakerBio:  Birgitte Jordal

We are four Norwegian women that hold a bachelor’s degree in Digital Infrastructure and Cybersecurity from NTNU. Our interests include CTFs, ethical hacking, and penetration testing, with experience in cloud infrastructure, secure networking, and threat analysis.

SpeakerBio:  Julia Kucharska
No BIO available
SpeakerBio:  Emilie Jørstad
No BIO available
SpeakerBio:  Selma Jenker
No BIO available

Creator: Radio Frequency Village

What does the fox say? Turns out the fox says quite a lot, you just need to know how to listen! A “fox” is a hidden transmitter, in this case a WiFi access point. In this talk we’ll go from basics to advanced fox hunting, covering: direction finding, antenna selection, channel scanning, refresh rate, remote sensors, mobile devices, staying stealthy so the fox never sees you, and working as a team. Not only is it a great DEFCON pastime, it’s a real-world skill that lasts long after the con.

People:
    SpeakerBio:  Eric Escobar, Sophos – Red Team Lead

Eric is a seasoned pentester and a Red Tech Lead at Sophos. On a daily basis he attempts to compromise large enterprise networks to test their physical, human, network and wireless security. He has successfully compromised companies from all sectors of business including: Healthcare, Pharmaceutical, Entertainment, Amusement Parks, Banking, Finance, Technology, Insurance, Military, Retail, Food Distribution, Government, Education, Transportation, Energy and Industrial Manufacturing.

His team consecutively won first place at DEF CON 23, 24, and 25’s Wireless CTF, snagging a black badge along the way. Forcibly retired from competing in the Wireless CTF, he now helps create challenges!

Creator: Cloud Village

Malware loaders used to advertise themselves with throwaway domains, fragile hosting, and obvious command-and-control infrastructure. Foxveil is a hard reset in loader tradecraft. Less attacker-owned infrastructure, less disk-heavy execution, fewer static clues, and more abuse of platforms defenders see every day. This newly documented loader stages payloads from our most trusted and favorite cloud-native tool chains: Cloudflare Pages, Netlify, and Discord attachments, executes shellcode largely in memory, establishes persistence through masqueraded Windows artifacts, and actively rewrites analysis-significant strings at runtime to make reverse engineering and static detection harder. Across two observed variants, Foxveil combines several pieces of modern tradecraft into one compact initial-stage loader. One variant spawns a fake svchost.exe and performs Early Bird APC injection into the target process before it fully resumes. Another pulls shellcode from Discord attachments and executes it through self-injection. Both stage follow-on payloads from trusted platforms, drop files into SysWOW64 under names designed to blend into normal Windows noise, and use runtime string mutation to corrupt keywords such as beacon, meterpreter, shellcode, and even fox itself. This talk walks through the full Foxveil infection chain, from staging and shellcode delivery to persistence and follow-on deployment, and examines what makes this loader part of a broader shift in attacker and malware tradecraft by comparing both variants. The real story is not just cloud-hosted staging. It is the emergence of a loader model that borrows trust instead of building infrastructure. Foxveil examines what modern initial-stage malware looks like when it is designed for rotation, ambiguity, and survival.

People:
    SpeakerBio:  Shani Kurtzberg

Shani Kurtzberg is an XDR Team Lead at Cato Networks and member of Cato CTRL. She leads the Threat Intelligence and XDR Detection Engineering initiatives. Prior to Cato, Shani served in the Israeli Air Force (IAF) as a Security Analyst, leading SOC operations to protect critical systems. Shani holds a Master of Business Administration (M.B.A.) from Peres Academic Center, specializing in Marketing and Product Management.

SpeakerBio:  Zohar Buber

Zohar Buber is a security analyst in Cato Research Labs at Cato Networks. He focuses on network protocol analysis and malicious traffic detection, specializing in threat identification using network-based methods. He previously worked at Radware, where he examined threats in the DDoS industry. Zohar holds B.A focused in Business Administration, Information System Analysis // Zohar Buber, zohar@devtalks.me

Creator: Car Hacking Village

Modern vehicles are increasingly adopting Automotive Ethernet for UDS-based diagnostics instead of CAN. While Gateway ECUs are designed to restrict communication to specific IPs, some support DHCP to facilitate communication with diagnostic tools. This DHCP support allows an attacker to deploy a rogue DHCP server via the OBD Ethernet interface, assigning a controlled IP to the Gateway ECU and enabling network-level attacks without any prior knowledge of the vehicle’s internal network. In this talk, we demonstrate that high-rate Layer 2/3 flooding — regardless of protocol (ARP, ICMP, UDP, TCP) — disrupts safety-critical systems including AVN, wipers, headlights, and causes abrupt stopping when shifted into Drive or Reverse, and that at higher packet rates the infotainment display blacks out and does not recover until fully powered down. We discovered these findings on a production vehicle and have prepared a demo.

Links:
    Github – https://github.com/LYHyoung

People:
    SpeakerBio:  Yehyeong Lee, Autocrypt Inc.

Yehyeong Lee is a cybersecurity engineer specializing in automotive security and embedded systems. He focuses on vulnerability analysis and penetration testing of vehicle ECUs and in-vehicle networks. His work includes real-world vehicle security testing and automotive network attacks. His research interests include automotive security, firmware analysis, and embedded system security.

Creator: Middle Easterns & Africans in Cyber Security (MEACS)

The word algorithm comes from a mathematician in ninth-century Baghdad. The first person to write down how to break a cipher was working in the Arab world more than a thousand years before modern computing. The foundations of the field we now call cybersecurity were laid by scholars from the Middle East and Africa, and that lineage runs straight through to the practitioners in this room. This session opens the MEACS Community Space with a look at those roots, from the origins of cryptanalysis and algebra to the engineers, defenders, and founders carrying the work forward today. We will talk about where we come from, why representation in this field is not a side conversation, and what this community is here to build over the next few days. Whether you trace your roots to the region or you just want the history the industry skipped, you are welcome here. Come meet the people who make this space what it is.

People:
    SpeakerBio:  Amber Bennoui
No BIO available
SpeakerBio:  Ezz Tahoun

Ezz Tahoun is an award-winning cybersecurity data scientist recognized globally for his innovations in applying AI to security operations.

He has keynoted, trained & presented at BlackHat US, Sector, MEA, Asia & EU, DEFCON, SANS Summits, all the top Bsides, Securityweek ICS Conference and GISEC among many others.

His groundbreaking work earned him a gold edison award and accolades from Yale, Princeton, Northwestern, NATO, Microsoft, and Canada’s CSE.

At 19, Ezz began his PhD in Computer Sci at the Univ of Waterloo, quickly gaining recognition through over 20 influential papers and open-source tools.

His experience includes leading advanced AI security ops projects for Orange CyberDefense, Forescout, RBC, and Huawei US.

He holds certifications such as GIAC Advisory Board, aCCISO, CISM, CRISC, GCIH, CEH, PMP and GCP-Cloud Architect, and served as an adjunct professor in cyber defense and warfare.

Creator: Red Team Village

Cloud-native teams often treat web app “config leaks,” verbose telemetry, and CI/CD misconfigurations as separate, low-impact issues. This talk shows how those assumptions break in practice. We present a full kill chain where a single exposed Application Insights key in a front-end patient portal is chained with realistic but common Azure misconfigurations to achieve complete tenant compromise and Global Administrator access.

Starting from an exposed Application Insights Instrumentation Key in client-side JavaScript, we demonstrate how an attacker can pivot into the Application Insights query API, mine telemetry for over-privileged SAS URLs, and use those tokens to harvest sensitive data from Blob Storage and Storage Queues. We then show how seemingly innocuous Function App endpoints, intended for document ingestion, can be abused via SSRF and URL-based command injection to access local files and environment variables, recovering additional credentials. Finally, we weaponize misconfigured Azure DevOps pipelines and Library Variable Groups to exfiltrate cloud credentials to a webhook, escalate privileges in Entra ID, and take control of production workloads.

Rather than introducing a single novel vulnerability, this session focuses on the combinatorial risk of real-world misconfigurations across Application Insights, Storage, Functions, DevOps, and identity. Attendees will walk away with concrete cloud hunting techniques, a step-by-step attack chain they can reproduce in their own labs, and a prioritized remediation playbook for breaking similar chains in their environments.

People:
    SpeakerBio:  Chirag Savla

Chirag Savla is a cyber security professional with 10+ years of experience. His areas of interest include penetration testing, red teaming, azure and active directory security, and post-exploitation research. For fun, he enjoys creating open-source tools and exploring new attack methodologies in his leisure. Chirag has worked extensively on Azure, Active Directory attacks and defense, and bypassing detection mechanisms. He is the author of multiple open source tools such as Process Injection, Callidus, and others. He has presented at many conferences and local meetups and has trained people in international conferences like Blackhat, BSides Milano, Wild West Hackin’ Fest, HackSpaceCon and VulnCon.

SpeakerBio:  Raunak “Trouble1” Parmar

Raunak Parmar works as a senior cloud security engineer at White Knight Labs with 6+ years of experience. His areas of interest include web penetration testing, Azure/AWS security, source code review, scripting, and development. He enjoys researching new attack methodologies and creating open-source tools that can be used during cloud red team activities. He has worked extensively on Azure and AWS and is the author of Vajra, AzDevRecon and MsCodePhish. He has spoken at multiple respected security conferences like Black Hat, Defcon, Nullcon, RootCon, HackspaceCon, NorthSec, LeHack , etc and also at local meetups.

Creator: Red Team Village

This presentation takes a practical, demo-driven approach to exploitation, demonstrating how a buffer overflow vulnerability can still serve as an entry point to compromise modern industrial infrastructures. The session is approximately 90% hands-on and 10% theoretical. The talk walks through the full attack chain, starting from initial exploitation and progressing to the development and execution of multistage shellcode. This shellcode establishes communication with a remote Command and Control (C2) server, retrieves an additional payload, and executes it directly in memory, bypassing traditional detection mechanisms. Once the environment is compromised, the session demonstrates how an attacker can gain full remote control, enabling command execution, lateral movement, and exfiltration of sensitive data, including critical assets such as licenses and proprietary industrial information. The talk also includes interaction with real-world devices such as RTUs (Remote Terminal Units), showing how malicious actions can directly impact industrial operations. In critical scenarios, this can lead to disruptions in essential services, including failures in energy systems and potential blackouts. The content is based on real-world pentesting experience conducted in an energy sector company in Brazil, providing a practical and applied perspective on risks in ICS/SCADA environments. The goal is to demonstrate how modern attacks combine vulnerabilities with advanced techniques, reinforcing the need for robust security strategies to protect critical infrastructure.

People:
    SpeakerBio:  Fernando Mengali, Information Security Specialist

Cybersecurity researcher focused on Pentesting and AppSec, also serving as a Practical / Hands-on Speaker. He also dedicates part of his research to critical infrastructure security (ICS/SCADA), exploring the intersection between classic vulnerabilities and industrial environments. He has over 18 years of experience in offensive security and practical application exploitation.

He has participated in research and development projects focused on vulnerability exploitation and offensive security.

Specialized in 0day discovery and exploit development, he has over 135 published CVEs and is ranked in the Top 10 contributors on VulDB https://vuldb.com/users.top.

Additionally, he has published multiple exploits and technical papers publicly available https://www.exploit-db.com/?author=12136 and https://packetstorm.news/files/author/8470/1.

His work focuses on real-world vulnerability exploitation, including advanced scenarios such as memory corruption, buffer overflows, and complex environments.

He is the creator of https://yrprey.com, an educational framework that brings together more than 20 vulnerable applications based on the OWASP Top 10, used for practical training in Application Security and Offensive Security https://owasp.org/www-project-vulnerable-web-application-directory.

He is also the driving force behind https://speakfy.io, a project focused on promoting Information Security events globally.

Furthermore, he develops application security-oriented tools, such as https://leapfix.co (static code analysis) and https://fitoxs.com, a dynamic application analysis platform powered by artificial intelligence.

SpeakerBio:  Thiago Cunha da Silva

Thiago Cunha is an offensive security specialist, or as he likes to put it: “professionally paid to break into systems before someone less friendly does.”

He currently works as a Red Team and Threat Intelligence Consultant at Banco Carrefour and as an instructor at the Kryfal, teaching future ethical hackers not to click on suspicious links.

Creator: AppSec Village

Most vulnerabilities aren’t found because teams lack tools they persist because the cost of fixing them is too high. A finding without a concrete, context-aware fix is just noise developers learn to ignore. This talk walks through three real-world attack scenarios a secret leaked into git history, a compromised dependency in a supply chain attack, and an injection vulnerability introduced in a PR and shows how each is detected, mapped to OWASP Top 10:2025, and automatically remediated using AI running entirely on local infrastructure. No source code leaves the machine. The AI receives the vulnerable code block, CWE identifier, CVSS score, and OWASP category and returns a fix that is specific, correct, and ready to apply. Attendees leave with a working open-source tool and a new mental model for what AppSec remediation can look like.

People:
    SpeakerBio:  Filipi Pires, Head of Technical Advocacy at SCYTHE

I’ve been working as Head of Technical Advocacy at SCYTHE, Founder & Investor at CROSS-INTEL, Advisor & Investor at Sherlockeye, BSides Porto Organizer, Red Team Village Director (DEF CON), Senior Advisor Raices Cyber Academy, Founder of Red Team Community (Brazil and LATAM), AWS Community Builder, Snyk Ambassador, Application Security Specialist and Hacking is NOT a crime Advocate. International Speaker at Security and New technologies events in many countries such as US (Black Hat & Defcon), Canada, France, Spain, Germany, Poland, Black Hat MEA – Middle-East – and others, I’ve served as University Professor in Master Degree in Portugal, Graduation and MBA courses at Brazilian colleges, in addition, I’m Creator and Instructor of the Course – Malware Attack Types with Kill Chain Methodology (PentestMagazine), PowerShell and Windows for Red Teamers(PentestMagazine) and Malware Analysis – Fundamentals (HackerSec).

Black Hat US 2025 – https://blackhat.com/us-25/arsenal/schedule/presenters.html#filipi-pires-46329 Black Hat US 2024 – https://blackhat.com/us-24/arsenal/schedule/presenters.html#filipi-pires-46329 Black Hat MEA 2025 – https://blackhatmea.com/speaker/filipi-pires-0 Black Hat MEA 2024 – https://blackhatmea.com/speaker/filipi-pires DEF CON 33 / 32 – https://sessionize.com/filipi-pires/ DEF CON – Adversary Village – https://adversaryvillage.org/adversary-events/DEFCON-33/Filipi-Pires/

Creator: Policy @ DEF CON

AI has upended vulnerability discovery. What used to be scarce is now abundant: vulnerabilities can be found faster, at greater scale, and across more systems than ever before. Systems that uplevel vuln hunters are becoming widely available, to both defenders and malicious actors. Finding vulnerabilities used to be the hard part – but it’s not anymore, and we need to adapt the vulnerability ecosystem to reflect that. But vulnerability management – thoughtful and methodical disclosure, triage, patching, and policy – still operates on assumptions of scarcity. That mismatch is becoming dangerous. If AI makes vulnerability discovery effectively infinite, what does a safe, scalable system for handling them actually look like? This panel will discuss how the vulnerability needs to evolve to match the moment. Current policy approaches – especially those mandating rapid reporting or broad sharing of unmitigated vulnerabilities – risk making systems less secure in an AI-driven environment. At the same time, rapid response by defenders is critical, and resources are scarce. Disclosure models need to evolve, governments need to support defenders, and incentives must align to ensure that AI-driven discovery strengthens security.

People:
    SpeakerBio:  Lindsey Cerkovnik, CISA

Lindsey Cerkovnik is the Chief of CISA’s Vulnerability Response & Coordination (VRC) Branch. Her team is responsible for CISA’s Coordinated Vulnerability Disclosure (CVD) process, the Known Exploited Vulnerabilities (KEV) catalog, and CISA’s Stakeholder Specific Vulnerability Categorization (SSVC) process. Lindsey and her team help to maintain, support, and advance the global vulnerability ecosystem by funding and overseeing the CVE and CVE Numbering Authority (CNA) programs, leading the production and dissemination of machine-readable vulnerability enrichment information, and engaging in valuable technical collaboration with the vulnerability research community.

SpeakerBio:  John Banghart, Center for Cybersecurity Policy & Law

John Banghart leverages his significant federal government and private sector experience in cybersecurity to navigate issues related to risk management, government policy, standards and regulatory compliance, and incident management. He has successfully led efforts to address significant and high-profile cybersecurity issues within major government programs and institutions while facing complex legal, technical, and political circumstances. From 2013 to 2015, John played a key role in developing the Obama administration’s cybersecurity and technology policy as the National Security Council’s director for federal cybersecurity. He led policy, technical, and process efforts to reduce cybersecurity risk and improve metrics and measurement for all civilian, military, and intelligence community agencies. He served as a primary advisor on cybersecurity incidents and preparedness and led the National Security Council’s efforts to address significant cybersecurity incidents, including those at OPM and the White House, among others. He also spent several years at the National Institute of Standards and Technology (NIST), both as a cybersecurity researcher and in the Office of the Undersecretary of Commerce for Standards and Technology. John also worked as a senior cybersecurity advisor for the Centers for Medicare and Medicaid Services, providing leadership to the cybersecurity preparations for the Healthcare.gov website.

SpeakerBio:  Ben Flatgard, JPMorgan Chase

Ben Flatgard is an Executive Director with JPMorgan Chase Co. He leads public policy development and advocacy, as well as partnership initiatives, to improve the cybersecurity of the firm, its customers and clients, and the broader digital ecosystem. Ben previously served in the Obama Administration from 2009-2017. In his most recent post as Director for Cybersecurity Policy on the National Security Council, Ben was responsible for leading cybersecurity policy development related to protection of critical infrastructure and emerging technologies. Before joining the National Security Council, Ben served as Senior Advisor to the Assistant Secretary of the Treasury for Financial Institutions. Prior to his time at Treasury, Ben held positions at the Department of Commerce and the White House. Ben graduated from the University of Edinburgh. He holds fellowships at the Atlantic Council in Washington, DC, and at the University of Sydney.

SpeakerBio:  Elizabeth Eigner, Microsoft

Elizabeth Eigner is a Senior Manager on Microsoft’s Global Cybersecurity Policy team, where she leads Microsoft’s vulnerability policy portfolio, overseeing efforts to develop and implement strategies that address vulnerability management both in the United States and globally. She represents Microsoft on the Hacking Policy Council, where she works collaboratively with industry leaders and policymakers to advance responsible cybersecurity and strengthen the frameworks that underpin software security worldwide. Elizabeth also leads initiatives aimed at creating and enhancing national cyber strategies in countries around the world. She works closely with governments and stakeholders to strengthen policy frameworks and promote resilient cybersecurity practices globally. Elizabeth also leads Microsoft’s Advancing Regional Cybersecurity (ARC) initiative, focusing on improving incident response capabilities and cyber capacity building in the Global South. Previously, she served as Microsoft’s representative on the Cloud Service Provider Advisory Board (CSP-AB), contributing to FedRAMP public policy discussions and best practices for cloud security. Before joining Microsoft, Elizabeth worked at The Washington Technology Industry Association to enhance Washington State’s innovation ecosystem. At MIT Solve, she collaborated with tech-based social entrepreneurs on solutions fostering digital inclusion and equitable economic opportunity. She holds a B.S. in Political Science from Northeastern University, with concentrations in Law and International Security.

Creator: Misc

A Reference Architecture for Modern Incident Response (in Rust)

Incident response still runs on a toolbox: a dozen single-purpose parsers, a pile of CSVs, a timeline stitched together by hand, and a senior analyst’s memory holding it all together. Every engagement re-solves the same plumbing. Every new artifact means a new script. And the hard-won expertise — how to read a prefetch run-count, how to resolve a ControlSet in an offline hive, which event IDs actually matter — lives in people, not in code you can run twice.

This talk argues for a different shape: incident response as a framework, not a toolbox — and backs the argument with a working reference implementation, Issen, an open-source DFIR engine written in Rust. You point Issen at evidence — an E01/EWF/VMDK or raw disk image, a memory dump — and it returns a single normalized timeline and an ATT&CK-mapped report. One static binary. No Python runtime, no agent, no cloud. It runs the same on an examiner’s laptop, in CI, or inside a sealed evidence enclave.

Under the hood, Issen is deliberately thin. The real work lives in a fleet of standalone, single-responsibility forensic libraries — each a deep expert in one artifact family (NTFS and the change journal, registry hives, prefetch, SRUM, browser history, EVTX, SQLite, Windows and Linux memory) — that know nothing about each other. Issen is the orchestration and correlation layer that wires them into one story and speaks one normalized vocabulary of findings (“consistent with,” never a verdict), each tagged to MITRE ATT&CK. We’ll walk the layered model that makes this composition work — container → filesystem / memory / log → parser → orchestration — and why every parser is medium-agnostic by design.

The most useful — and least glamorous — lesson is the one the talk is really about: capability is cheap to build and expensive to wire. Demonstrated live against the public “Stolen Szechuan Sauce” case, we’ll show artifacts that were completely correct in isolation yet invisible end-to-end, because a decoder existed but nothing called it, or a file was extracted but never classified, or a real-world quirk (offline hives have no CurrentControlSet; large registry values are split into “big data” segments) quietly returned nothing. We’ll trace each from “zero results” to ground truth.

Engineering substance throughout: panic-free, forbid(unsafe) parsers that treat every input as hostile; correctness validated against independent external oracles (not just the authors’ own fixtures); and a clean separation between observed fact, forensic inference, and the conclusions that belong to a human, not a tool.

You’ll leave with: a concrete, composable architecture for an IR pipeline you can adopt or fork; a working open-source reference to start from today; a repeatable method for finding “dark” capability in your own tooling; and a sober view of what to automate, what to keep human, and how to phrase findings so they survive scrutiny.

People:
    SpeakerBio:  HUI Kwun Tai, Albert (4n6h4x0r), DC852

Albert Hui (@4n6h4x0r) is a digital forensics and incident response practitioner and the founder of Security Ronin. He builds open-source forensic tooling — including Issen, a Rust DFIR engine, and a family of panic-free libraries spanning disk, memory, registry, and application artifacts — and works hands-on as an examiner and expert witness testifying before courts of law, where findings have to survive cross-examination, not just look good on a slide. He’s drawn to tooling an analyst can actually run twice: reproducible, offline, and honest about the line between observed fact and inference. In a past life he was an IBM Global Security Architect and a Deloitte Risk Advisory Director — which goes some way to explaining his propensity for horizontal thinking.

Creator: Red Team Village

Kiosks are everywhere. Their prevalence in our society has exploded, with devices such as fast-food ordering, signage, ticketing and check-in, print/ship/office, library, point-of-sale, and more – all around us.

How secure are they though? Are they being adequately tested? We all learned on July 19, 2024, when a multitude of the signs and kiosks around us went BSOD that a large number of these kiosks – from airports to ATMs – are running on Windows. We need to be asking questions like, are they joined to a domain, what is network adjacent, and how could these systems lead to a serious attack path?

In this workshop we will walk through how to hack and secure kiosks in a four-part workshop, which will move from initial compromise to lateral movement and domain compromise, concluding with some defensive recommendations to prevent ‘kiosk to domain takeover’ scenarios.

People:
    SpeakerBio:  Ezra Woods

Avid security researcher currently working as an Information Security Engineer with Grand Canyon Education

SpeakerBio:  Mike Manrod

Mike, AKA, CroodSolutions, presently serves as CISO for Grand Canyon Education, teaches Malware Analysis for GCU, and helps lead the Threat Intelligence Support Unit.

In addition to work on BypassIT, AutoPwnKey, BYOEDR, and BeaconatorC2 with Ezra Woods and other contributors, Mike has served as a threat prevention engineer for Check Point along with various other roles.

He is also a co-author/contributor for the joint book project, Understanding New Security Threats published by Routledge in 2019, along with multiple articles/whitepapers. When not working, he spends time playing video games or working on cars with his kids.

SpeakerBio:  Spencer Alessi

Spencer is a passionate security practitioner, with the heart of a defender and spirit of a hacker. Spencer’s background is in IT and Systems Administration prior to making the transition to security. He leans on these crucial technical skills to help enable clients and security programs to build defensible, vigilant, and practical security programs. Most of Spencer’s current efforts are focused on understanding threats, techniques, and methods and then implementing them through technical assessments and analysis. Spencer is Co-Host of The Cyber Threat Perspective podcast and regularly creates blogs, videos and other content. In November 2025 Spencer was awarded Microsoft MVP in Security – Identity & Access.

Creator: La Villa Community

The best hiding place is often a binary Microsoft already signed. This talk presents a complete red team operation reconstructed from a real multi-stage JScript dropper captured in April 2026 targeting Energy, Government, and Aerospace organizations. Using custom tooling and malware analysis, we transformed the adversary’s workflow into a repeatable offensive playbook focused on trusted binary abuse, fileless execution, and stealthy process injection.

The session demonstrates how attackers delivered a .NET payload through steganographic C2 hosted on trusted paste services, loaded assemblies directly in memory with Reflection.Assembly::Load(), and performed Process Hollowing into msbuild.exe using native Windows APIs including ZwUnmapViewOfSection, VirtualAllocEx, and SetThreadContext.

We also explore operational evasion techniques such as WMI hidden process spawning, obfuscated Base64 transformations, and Sysmon telemetry analysis from the perspective of a red team operator. Every technique shown was extracted from a live adversary sample and operationalized for realistic adversary emulation.

People:
    SpeakerBio:  Filipi Pires, Head of Technical Advocacy at SCYTHE

I’ve been working as Head of Technical Advocacy at SCYTHE, Founder & Investor at CROSS-INTEL, Advisor & Investor at Sherlockeye, BSides Porto Organizer, Red Team Village Director (DEF CON), Senior Advisor Raices Cyber Academy, Founder of Red Team Community (Brazil and LATAM), AWS Community Builder, Snyk Ambassador, Application Security Specialist and Hacking is NOT a crime Advocate. International Speaker at Security and New technologies events in many countries such as US (Black Hat & Defcon), Canada, France, Spain, Germany, Poland, Black Hat MEA – Middle-East – and others, I’ve served as University Professor in Master Degree in Portugal, Graduation and MBA courses at Brazilian colleges, in addition, I’m Creator and Instructor of the Course – Malware Attack Types with Kill Chain Methodology (PentestMagazine), PowerShell and Windows for Red Teamers(PentestMagazine) and Malware Analysis – Fundamentals (HackerSec).

Black Hat US 2025 – https://blackhat.com/us-25/arsenal/schedule/presenters.html#filipi-pires-46329 Black Hat US 2024 – https://blackhat.com/us-24/arsenal/schedule/presenters.html#filipi-pires-46329 Black Hat MEA 2025 – https://blackhatmea.com/speaker/filipi-pires-0 Black Hat MEA 2024 – https://blackhatmea.com/speaker/filipi-pires DEF CON 33 / 32 – https://sessionize.com/filipi-pires/ DEF CON – Adversary Village – https://adversaryvillage.org/adversary-events/DEFCON-33/Filipi-Pires/

Creator: Telecom Village

1. IPTV Architecture – How IPTV is delivered over FTTH/GPON and the security implications of ONTs and STBs.
2. Weak Links – Common vulnerabilities in ISP-managed ONTs.
3. ONT to STB Attack Chain – A real-world red team attack demonstrating remote compromise and IPTV manipulation.
4. Telecom SOC Detection – Detecting ONT compromises through management, authentication, configuration, and IPTV traffic monitoring.

People:
    SpeakerBio:  Zibran Sayyed
No BIO available

Creator: Red Team Village

In this hands-on tactic, attendees will exploit a realistic document management portal to chain path traversal into something far more dangerous. We’ll demonstrate a complete attack path from initial discovery to domain credentials, then discuss why this works, where you’ll find it, and how defenders should be looking for it.

What You’ll Do: You’ll get hands-on access to a vulnerable Windows IIS application and work through the complete exploitation chain. By the end of the session, you’ll have captured real domain credentials and understand exactly how the attack works at each step.

Basic familiarity with web vulnerabilities helpful but not required. We’ll explain the concepts as we exploit them.

People:
    SpeakerBio:  Mike Lisi

Mike helps organizations identify, prioritize, and eliminate real-world security risks before attackers exploit them. As Founder & CEO of Maltek Solutions, he brings over 15 years of hands-on experience across penetration testing, risk assessments, and cyber capability development.In addition to his consulting leadership, Mike serves as President of Red Team Village, a 501(c)(3) nonprofit delivering free offensive security training to a global community. Mike also leads CTF design for NCAE CyberGames, engaging 1,000+ collegiate participants annually.

Creator: Cloud Village

Azure DevOps service connections are a core part of modern pipelines, allowing teams to interact with Azure and other external systems seamlessly. While they make deployments faster and easier, they also introduce an attack surface that is often overlooked in real-world environments.

In this research, we explore new attack paths showing how even limited access within Azure DevOps pipelines can be leveraged to escalate privileges and compromise cloud environments. By taking advantage of over-scoped service connections and certain platform behaviors, an attacker can chain together small gaps to achieve significant impact, including gaining subscription-level access.

These findings were reported to Microsoft and acknowledged through MSRC, highlighting a deeper issue in how DevOps integrations are secured. Rather than being just a configuration problem, this reflects a broader gap that organizations need to understand when designing their cloud security approach.

People:
    SpeakerBio:  Saksham Agrawal

Saksham Agrawal is a Senior Security Consultant at NotSoSecure, specializing in cloud security. His work focuses on discovering new attack paths in cloud environments and helping organizations understand real-world risks. He has presented his research at DEF CON Cloud Village, where he introduced his tool NoPrompt and shared practical techniques for cloud security testing. He enjoys building tools, exploring cloud internals, and sharing his findings with the security community. He has also responsibly reported critical vulnerabilities in major cloud vendors as part of his independent security research and actively delivers training sessions on cloud security and offensive security techniques.

Creator: Policy @ DEF CON

Every frontier AI lab publishes policies, but far less is known about what happens between a line in that policy and a blocked request in production. On Anthropic’s safeguards team, we operate in that gap. In this talk, we’ll walk through the full enforcement lifecycle at Anthropic: how policy gets written, how we translate it into evals that actually measure the behavior we care about, how we build precision-tuned classifiers to detect misuse traffic at scale, and how those classifiers feed a layered enforcement stack. We’ll share what’s worked, what hasn’t, and the open problems we think the policy community should be paying attention to.

People:
    SpeakerBio:  Grant Versfeld, Anthropic

Grant Versfeld co-leads cyber enforcement on Anthropic’s Safeguards team, where he bridges the Policy and Threat Intelligence teams to disrupt cyber threat actors. Before Anthropic, he was a security engineer for Google Cloud Threat Intelligence, investigating nation-state campaigns against Cloud customers, contributing to Google Cloud’s Threat Horizons report, and supporting Mandiant’s cyber espionage research. He is a former Cybersecurity Policy Fellow at the Center for Democracy & Technology, a Hackers on the Hill ambassador, and a judge for Atlantic Council’s Cyber 9/12 competition. He returns to DEF CON having previously played the FBI agent in DC 31’s Hacker Court.

SpeakerBio:  David “0xdf” Forsythe, Anthropic

David Forsythe (0xdf) is a member of technical staff at Anthropic working on cybersecurity strategy and safeguards for frontier AI. Before that, 20+ years across the US public and private sectors in SOC/CIRT, threat intelligence, red teaming, threat research, and education, most recently as Principal Lab Architect at HackTheBox.

He is also a long-running security content creator, publishing CTF walkthroughs and technical breakdowns of exploits and other security topics at 0xdf.gitlab.io and on YouTube. As a player, he has reached as high as 4th on the HackTheBox global leaderboard, is a two-time Best Overall winner of the SANS Holiday Hack Challenge, and a three-time SANS Netwars Tournament of Champions team winner.

Creator: AppSec Village

Users building apps with vibe coding platforms like Lovable or Base44 often overlook critical security vulnerabilities. As these platforms gain rapid adoption, thousands of apps are being created without sufficient attention to security, leaving significant flaws in plain sight. Our research, conducted across thousands of apps, revealed widespread issues, including 175 instances of sensitive PII leakage (such as medical records and personal contact information), SSRF, 0-click account takeovers, and IDORs.

In this talk, we’ll share our field experience in developing discovery and security testing techniques that uncover exploitable web app and API behaviors at scale. We’ll walk through concrete examples, showing how critical vulnerabilities hide in plain sight and how even minor misconfigurations can lead to catastrophic breaches. Attendees will learn to identify these security risks and secure their no-code apps, including best practices for handling sensitive data and securing APIs.

People:
    SpeakerBio:  Samantha Pearlstein, Founding Solutions Engineer at Escape

Samantha is a solutions engineer with a strong background in security research, executive cyber resilience, and nation-state threats. As a former consultant at Accenture, she led cyber resilience initiatives for Fortune 100 executives, developed GenAI-powered security tools, and delivered workshops on emerging cyber challenges. Today, as a Sales Engineer at Escape, Samantha helps AppSec teams secure their APIs and SPAs, combining her passion for cybersecurity with hands-on problem-solving.

Creator: Middle Easterns & Africans in Cyber Security (MEACS)

If yesterday was where we came from, today is where we are going. The same regions that gave the world algebra and the first codebreakers are now home to some of the fastest-growing security communities on the planet, from Lagos to Cairo to Riyadh, alongside a diaspora shaping the field from inside every major company and conference. This talk picks up where the kickoff left off and moves from history to right now: the researchers, founders, defenders, and organizers carrying the work forward, the obstacles that still stand in the way, and what it takes to build a career and a community when the industry has not always made room for you. We will talk honestly about representation, mentorship, and momentum, and about the role a space like MEACS plays in turning a shared heritage into a shared future. Come for the story, stay to find your people.

People:
    SpeakerBio:  Ezz Tahoun

Ezz Tahoun is an award-winning cybersecurity data scientist recognized globally for his innovations in applying AI to security operations.

He has keynoted, trained & presented at BlackHat US, Sector, MEA, Asia & EU, DEFCON, SANS Summits, all the top Bsides, Securityweek ICS Conference and GISEC among many others.

His groundbreaking work earned him a gold edison award and accolades from Yale, Princeton, Northwestern, NATO, Microsoft, and Canada’s CSE.

At 19, Ezz began his PhD in Computer Sci at the Univ of Waterloo, quickly gaining recognition through over 20 influential papers and open-source tools.

His experience includes leading advanced AI security ops projects for Orange CyberDefense, Forescout, RBC, and Huawei US.

He holds certifications such as GIAC Advisory Board, aCCISO, CISM, CRISC, GCIH, CEH, PMP and GCP-Cloud Architect, and served as an adjunct professor in cyber defense and warfare.

Creator: Adversary Village

Knowing who is coming for you is only half the fight. The real work is turning that knowledge into defenses that actually works. This panel walks through the full playbook, from raw threat intelligence on state-sponsored actors and other adversaries, to emulating their tradecraft, to validating whether your defenses can really stand up to it. Our panelists bring hands-on experience across adversary simulation, threat intel, and purple teaming to talk about what it takes to move from a report on paper to a tested, measurable defensive posture. The panel will dig into how to prioritize the threats that matter to your organization, how to faithfully replicate real adversary behavior instead of chasing generic checklists, and how red and blue working side by side turns every simulated attack into a lasting defensive victory. Whether you are defending against nation-state operators or opportunistic criminals, the goal is the same! Stop guessing about your defenses and start proving them. Join us for a practical conversation about closing the gap between intelligence and action.

People:
    SpeakerBio:  Cheryl Biswas, Strategic Threat Intel Analyst at TD Bank

Cheryl Biswas is a Strategic Threat Intel Analyst with TD bank in Toronto, Canada. She found her way into InfoSec through a helpdesk backdoor and pivoted into roles for vendor and change management, jumped a gap into privacy and DR/BCP, then laterally moved into security audits and assessments. Her degree in Political Science has evolved into researching APTs, botnets, ransomware and more. Cheryl is actively involved in the security community as a conference speaker and volunteer, mentors those entering the field, and encourages women and diversity in Infosec as a founding member of the “The Diana Initiative.”

SpeakerBio:  Adam Pennington, ATT&CK lead at MITRE Corporation

Adam Pennington leads ATT&CK at The MITRE Corporation and collected much of the intelligence leveraged in creating ATT&CK’s initial techniques. He has spent much of his 16 years with MITRE studying and preaching the use of deception for intelligence gathering. Prior to joining MITRE, Adam was a researcher at Carnegie Mellon’s Parallel Data Lab and earned his BS and MS degrees in Computer Science and Electrical and Computer Engineering from Carnegie Mellon University. Adam has presented and published in several venues including FIRST CTI, USENIX Security, DEF CON, and ACM Transactions on Information and System Security.

SpeakerBio:  Olaf Hartong, Defensive Specialist at FalconForce

Olaf Hartong is a Defensive Specialist and security researcher at FalconForce. He specialises in understanding the attacker tradecraft and thereby improving detection. He has a varied background in blue and purple team operations, network engineering, and security transformation projects. Olaf has presented at many industry conferences including Black Hat, DEF CON, DerbyCon, Splunk .conf, FIRST, MITRE ATT&CKcon, and various other conferences. Olaf is the author of various tools including ThreatHunting for Splunk, ATTACKdatamap and Sysmon-modular. He maintains a blog at https://olafhartong.nl

SpeakerBio:  Sarah Hume, Purple Team Service Lead at Security Risk Advisors

Sarah leads the Purple Team service at Security Risk Advisors (SRA). She has led hundreds of Threat Intelligence-based Purple Team exercises for organizations in the Fortune 500 and Global 1000 over the past 7 years. Her background is in offensive security, primarily internal network, OT/ICS, and physical security penetration testing. Sarah also has experience in external network penetration testing, web application assessments, OSINT, phishing/vishing campaigns, vulnerability management, and cloud assessments. Sarah graduated Summa Cum Laude from Penn State with a B.S. in Cybersecurity. She is a Certified Red Team Operator (CRTO), Certified Information Systems Security Professional (CISSP), Google Digital Cloud Leader, AWS Certified Cloud Practitioner, and Advanced Infrastructure Hacking Certified. She lives in Philadelphia with her dog, Paxton.

Creator: Telecom Village
Details:

Frontier AI is changing the economics and speed of cyberattacks—compressing vulnerability discovery, exploitation, and attack automation from days to potentially hours or minutes. For telecom networks and their complex OEM and supply-chain ecosystem, traditional vulnerability management and defense models may no longer be fast enough.

This session explores how telecom operators, OEMs, security teams, and regulators must adapt for the frontier AI era—from AI-driven vulnerability discovery and autonomous attacks to faster remediation, AI-powered defense, supply-chain resilience, and coordinated industry response. The key question: when attackers operate at machine speed, can telecom defense keep up?

People:
    SpeakerBio:  Arvind Singh
No BIO available

Creator: AppSec Village

uggingFace has become the GitHub of machine learning. Two million models, 15 million daily downloads. We discovered a critical RCE (CVE-2026-4372) that compromises any machine loading a malicious model. Add one field to a model’s config.json, publish it on the Hub, and any standard from_pretrained() call is silently compromised. No trust_remote_code=True. No warnings. The one security boundary the entire ML ecosystem relies on, bypassed completely.

The chain is three flaws: a generic setattr loop stamping every JSON field onto the config object including private attributes; a sanitization gap protecting the public interface but not the underscore-prefixed internal equivalent; and an unsandboxed kernel loader executing arbitrary Python from any Hub repo. One field weaponizes all three.

We walk through the full attack chain live. Model registries are the new package registries, facing the same supply-chain threats that broke the software world – with far less mature defenses.

People:
    SpeakerBio:  Yotam Perkal

Yotam Perkal leads security research at Pluto Security, a next-generation AI security and governance platform designed to protect the rapidly emerging ecosystem of AI builders, low-code/no-code tools, and agentic applications. His work focuses on securing AI-native development environments and building scalable methods for detecting, validating, and mitigating risks in AI-driven software workflows.

Previously, Yotam led the Threat Research team at Zscaler, headed the Vulnerability Research team at Rezilion, and held multiple roles within PayPal’s security organization across vulnerability management, threat intelligence, and insider threat.

Yotam is an active participant in several cross-industry working groups dealing with AI security, vulnerability management, and supply chain security.

Creator: OWASP Foundation

Join us for a state of Veilid update, where we’ll explore the latest progress, milestones, and what’s next for the project as it continues to build a decentralized, privacy-first communications and networking platform.

After the presentation, stick around for an open discussion about what it really takes to escape today’s digital walled gardens. We’ll talk about the challenges of reducing dependence on centralized services, the practical realities of building and adopting self-hosted and decentralized alternatives, and how open source communities can create tools that preserve privacy, autonomy, and user ownership.

People:
    SpeakerBio:  Gibson
No BIO available
SpeakerBio:  medus4
No BIO available
SpeakerBio:  cdC & Veilid crew
No BIO available

Creator: Bug Bounty Village

Every year the BBV badge gets one new thing we don’t know how to do yet. Last year: epoxy. This year: a real 0.96″ OLED behind the acrylic, giving a whole new vibe to the hacker laptop screen. Plus RGB for whichever team you’re on, and a fun CTF.

People:
    SpeakerBio:  Abhinav “TweetsFromPanda” Pandagale, Founder, Hackerware

Abhinav’s artistry comes from the times he used to sneakily paint drawings made by his sister. His hacking career began as a toddler, disassembling his toys but never putting them back together. His entrepreneurial roots come from selling snacks at a school fair and making a loss of ten bucks, his entire pocket money. Having learned how not to make money, he launched Hackerware.io – a boutique badgelife lab with in-house manufacturing – which has grown over the past ten years into a global presence across 20+ countries. He’s often spotted at conferences around the world – hosting hardware villages or pulling off the kind of random shenanigans that earned him the Sin CON Person of the Year 2025 award.

Creator: Hackers.town

People:
    SpeakerBio:  Janet Vertesi, Officer at Tech Reclaimers
No BIO available

Creator: Game Hacking Village

This informal talk is intended to cover expert usage of AI agents using Claude/Codex and how you can apply them to be effective in areas like game hacking. How do you keep an agent on track, how do you make it more accurate, more efficient, and how do you get it to go around guiderails. If you’re using AI in a web browser or copy+pasting code (in game hacking or otherwise) this talk is for you.

Links:
    Website – https://gamehacking.gg

People:
    SpeakerBio:  Juno, Game Hacking Village
No BIO available

Creator: Noob Community

This session introduces attendees to the MITRE ATT&CK Defense Game, a technical capstone project designed to bridge high-level gaming logic with real-world cybersecurity infrastructure. We will explore how gamification can be used to visualize adversary tactics and techniques in a controlled, interactive environment. The presentation covers the development of frontend interfaces and backend logic used to simulate defensive responses to various attack vectors, making complex security frameworks accessible to those just starting their journey in the field.

People:
    SpeakerBio:  Annie Meaney
No BIO available
SpeakerBio:  Anne Drago
No BIO available

Creator: AppSec Village

The tale of CVE-2026-40412, hacking Microsoft Planetary Computer Pro service. This talk will walk you through the journey of discovering a chain of vulnerabilities in the service itself and in its open source dependencies. All those findings combined resulted in cross tenant access. From a file read, to exploiting a buffer overflow (bypassing ASLR) to finding and exploiting a devastating design flaw. This talk will address the unique challenges of exploiting memory corruption bugs in cloud services, the pitfalls of cloud services’ architecture and how we helped the engineering team fix the issues we found at the design level.

People:
    SpeakerBio:  Michal Kamensky

Michal is a security researcher on the Microsoft STORM team, where she focuses on vulnerability research across the hybrid cloud domain. Previously she has worked as a security researcher at Bounce Security, and for the last few years volunteers as a Defcon goon. She enjoys diving into large code bases, understanding complex architectures, and eliminating vulnerabilities at scale.

Creator: OSINT For Good Community

What happens when you hand students a daily “Cyber Briefing,” a target, and unrestricted internet access? You build relentless investigators. In this talk, former Oracle vulnerability analyst turned educator Ben Crenshaw breaks down how to weaponize curiosity by scaling OSINT training from a single classroom to massive regional competitions. Drawing on his experience training 1000+ competitors for the Mayors Cyber Cup and coaching TraceLabs SearchParty CTFs, Ben will share exactly how to teach the threat-intel mindset. Discover how gamified open-source intelligence is reshaping cybersecurity education, turning passive scrollers into privacy-aware analysts, and building the next generation of defenders.

People:
    SpeakerBio:  Ben “The Cyber Sensei” Crenshaw, Transofotech Academy | Cyber & AI Education Leader | EdTech & Workforce Development | U.S. Cyber Games Mentor | NICE Cybersecurity Ambassador | CLEAR AI Initiative | Senior TraceLabs Coach | Aerospace Education Member

Ben Crenshaw, known as The Cyber Sensei, is a cybersecurity educator, author of two industry books, and a dedicated US Cyber Games Mentor. Leveraging his background as a former Senior Vulnerability Analyst at Oracle, Ben transitioned into education to bridge the critical gap between real-world threat intelligence and classroom theory. Currently serving as Lead Cybersecurity instructor for Transfortech and as former Head of Cyber and AI Education at Work ED, he has taught Open-Source Intelligence (OSINT) to thousands of students, shaping them into ethically grounded, privacy-conscious defenders. A passionate advocate for scaling cyber education, Ben actively trains educators and coaches to prepare the next generation of talent for large-scale competitions like the Mayors Cyber Cup and TraceLabs.

Creator: Noob Community

Industrial Control Systems (ICS) and Operational Technology (OT) run the world around us. Power plants, offshore oil rigs, trains, and other transportation systems, manufacturing plants – these are just a few examples of the critical infrastructure that society depends on. Each ICS/OT environment is unique and has specialized security requirements.

Protecting critical infrastructure becomes increasingly important each day as the frequency of cyberattacks and the number of attackers continue to grow. State adversaries are no longer the only ones targeting these specialized environments. Today’s attackers include ransomware groups, hacktivists, cyber mercenaries, and more.

ICS/OT cybersecurity can seem complicated and even daunting at first, but it does not have to be. This session will help participants understand the fundamentals of how these environments operate and how to secure them with a practical, simple approach.

People:
    SpeakerBio:  Mike Holcomb
No BIO available

Creator: Ham Radio Village

Radio Direction Finding (RDF) is one of the most practical and fun skills in amateur radio. This intro talk covers everything a beginner needs to know: low cost gear setups, building a simple Yagi antenna, and the techniques used to zero in on a transmitter.

Entry level talk reviewing what foxhunting / radio direction finding is Covering the basic equipment requirements Yagi antenna building Body fading Minimum equipment Tips and tricks

People:
    SpeakerBio:  Nate Moore

Nate Moore works in cybersecurity, holds an Extra class amateur radio license, and spends his spare time teaching, examining, and pushing signals around the ether.

Creator: Packet Hacking Village

LinkedIn hosts over 1 billion profiles and a growing number are fake. From AI-generated profile photos to synthetic work histories, adversaries use fake identities for spear-phishing, social engineering, and influence operations. This talk breaks down the anatomy of a fake LinkedIn profile and demonstrates an AI-powered detection framework using behavioral signals, image forensics, and network graph analysis. Attendees will leave with a hands-on methodology and open-source toolset to identify synthetic identities before they become insider threats. Prior knowledge of networking fundamentals is helpful; no machine learning background is required.

People:
    SpeakerBio:  Aiswarya Venkitesh, Principal Cloud Solution Architect, Microsoft

Aiswarya Venkitesh is a Principal Cloud Solution Architect at Microsoft Canada with 13+ years in Data & AI, specializing in agentic AI architecture, multi-agent orchestration, and enterprise Azure deployments. Ranked #4 globally in IT & Tech by Favikon with 55,000+ LinkedIn followers, she studies identity authenticity and influence operations on professional networks as both a practitioner and creator. She is a confirmed keynote speaker at Futura AI Conference 2026 and author of the forthcoming book The Agentic AI Playbook.

Creator: Recon Village

People are getting hired and trusted every day. Some of them do not exist at all, yet they still pass interviews, collect paychecks, and gain access to sensitive systems. Campaigns attributed to the DPRK have shown that this threat is very real. So how do you catch a ghost with a resume? Attendees will learn practical OSINT techniques for spotting fake personas and receive a checklist for thorough background checks. They will see these methods applied through two cases based on a true story, illustrating how these personas succeeded, how one could have been prevented, and where OSINT reaches its limits. These techniques not only help attendees detect fake personas but also provide practical ways to protect their own privacy and control what personal information is visible online.

People:
    SpeakerBio:  Michael Reimsbach, Product Security Specialist at SAP

Michael is a Product Security Specialist at SAP, working with the SAP Cloud Infrastructure security team. His focus areas include vulnerability management, secrets management, and building secure internal services.

He obtained multiple industry certifications such as OSCP, GCPN, and CISSP.

A healthy dose of paranoia led him to explore OSINT and the surprising power of publicly available information.

Beyond his day-to-day work, Michael is an active member of the cybersecurity community and helps organize BSides Luxembourg.

SpeakerBio:  Rishi “rxerium” C, Security Researcher

Rishi is a London-based security researcher with over five years of hands-on experience in IT. He currently specializes in vulnerability research, threat intelligence, and enterprise risk analysis. His current focus lies in identifying and analyzing zero-day vulnerabilities and emerging CVEs, often working to reverse engineer exploit mechanics and build detection logic before public weaponization. Rishi’s work spans both offensive and defensive domains—developing threat models based on real-world TTPs, crafting custom detection rules, and automating reconnaissance pipelines to uncover exploitable misconfigurations and exposed assets. He is particularly active in attack surface management (ASM) and OSINT, where he leverages DNS enumeration, passive data correlation, and large-scale infrastructure scanning to surface unknown entry points and map adversary-accessible exposure. Outside of research, Rishi integrates findings into operational tooling and supports data-driven prioritization strategies to bridge technical risk and business impact. His work reflects a deep commitment to adversary-informed defense and proactive discovery across modern hybrid environments.

Creator: Red Team Village

Critical infrastructure protocols designed in the late 70s were never built to handle a modern adversary. In this hands-on Tactic, we move beyond theoretical slides to execute a high-fidelity, nation-state-style operation against a municipal water treatment facility. Using a portable Software-in-the-Loop (SiL) environment, participants will experience the raw tradecraft of industrial sabotage without the need for a six-figure hardware rack.

This session focuses on “Living-off-the-Ladder” (LotL)—the art of using legitimate industrial function codes to perform unauthorized actions. You will not be dropping malware; you will be programmatically manipulating the physics of the plant using the same TTPs seen in recent geopolitical conflicts.

The Mission (30–60 Minute Sprint):

Industrial Interrogation: Use unauthenticated DNP3 and S7comm discovery to fingerprint PLCs and map the process logic—capturing vendor data and firmware versions silently.

The Memory Heist: Execute a “Credential Harvest” (T0861) by programmatically reading PLC holding registers to recover plaintext maintenance keys.

Coordinated Sabotage: Trigger a “Slow-and-Low” chemical setpoint override (Modbus FC16) across multiple facilities simultaneously, observing real-time physical feedback from the simulated plant sensors. • Phase 1: Automated multi-protocol fingerprinting and asset discovery. • Phase 2: Executing a “Memory Heist” to extract plaintext maintenance credentials directly from PLC registers. • Phase 3: A coordinated “Slow-and-Low” chemical setpoint override, providing real-time visual feedback of the physical process failure.

Attendee Takeaway: Participants will build the muscle memory required to navigate OT environments and will leave with a deployable, containerized SiL framework to continue their own ICS research and detection validation at home. Attendees will leave with a deployable SiL framework and the “Living-off-the-Ladder” playbook to continue their own ICS research without the need for expensive hardware.

People:
    SpeakerBio:  Blessen Thomas

Blessen Thomas is an Independent Security Researcher.He has more than 13+ years of experience in Red Teaming, Appsec (Web, Thick, API & Mobile Apps), Smart Watch Wearable Application Penetration Testing, Mobile Penetration Test (iOS,Android,Windows platform), IoT,OT ,mainframes,SAP,SWIFT,RPA,Cloud,ATM,KIOSK,Vulnerability Assessment and Network Penetration Test,Physical Covert Entry,Wireless assessments,Telecom(2G,3G,USSD) etc. for several enterprise companies and financial institutions all across the globe. He is a B.Tech in Information Technology from Anna University and holds industry certifications such as SANS GPEN,CRTO,OPST,CREST CRT(PEN),CREST CPSA,OSCP,CRTP,OSWP,C)PTE,CEH,CHFI. He has been listed and acknowledged in various “HALL OF FAMES” for various companies such as Oracle,Sony, Kayako, Appcelerator, Hotgloo, Meldium, Splunk and many more for responsible disclosure. He has been a bug bounty hunter and contributor for the OWASP Mobile Testing Guide Project(MSTG),Tamer OS, Seclists, OWASP top 10 API, OSSTMM, Awesome Mainframe Hacking,RvR,WAVSEP Benchmark sectool projects. His research training/talks has been accepted into various security conferences like Hack in the Box-Dubai,Hacktivity -Hungary, CanSecWest -Canada,OWASP Appsec EU- London -UK,OWASP Appsec Europe-Italy, RootCon-Philippines ,OWASP PH,OWASP New Zealand Day, Infosec SouthWest,Austin,Texas, FSec-Croatia, Hackbeach, Hackfest, Shakacon,ITWeb-South Africa, Jordan Cyber Security Summit, HITCON-Taiwan, OWASP AppSec-Bucharest, OWASP Appsec Africa-Morocco,CircleCityCon,OWASP Botswana,CactusCon,Bsides-London,Prishtina,Aarhus,Vilnius,Elbsides,Kristiansand,Athens and many more. He has been invited as Speaker for Radio Talk Shows for All India Radio. He spends his leisure time playing drumkit and percussion.

SpeakerBio:  Javier Hernández

Javier is a Red Team Operator and Malware Developer specializing in offensive engineering, adversary emulation, and custom tooling development. Holding certifications such as OSEP and CRTO, he has delivered high‑impact security engagements across both consulting environments and in‑house security teams. His work focuses on crafting stealthy implants, evasion techniques, and automation‑driven offensive capabilities. Passionate about applied research, he explores the intersection of malware development and AI‑augmented offensive security to help organizations strengthen their resilience against modern threats

SpeakerBio:  Wojciech Poparda

Wojciech Poparda is a cybersecurity consultant. He helps organizations proactively defend their digital infrastructure by thinking like an adversary. Leveraging a deep foundation in Offensive Security, he specializes in designing resilient Security Architectures that bridge the gap between complex attack vectors and enterprise defense, with a sharp focus on Cloud Security and Identity & Access Management (IAM).

His approach is backed by rigorous technical validation, holding industry-leading certifications including OSCP, CRTE, CRTP, CRTO, CPTS, CWES, AWS Certified Solutions Architect – Associate and AWS Certified Security – Specialty. He is also an Associate of ISC2, having successfully passed the CISSP exam.

Beyond the terminal, he channels the discipline, focus, and resilience required for cybersecurity into his life as a triathlete. As an IRONMAN European and World Championships finisher, he brings the same endurance and dedication to solving complex security challenges as he does to the race course.

Creator: Cloud Village

This is not another subdomain takeover 101 talk. It’s about what dangling DNS actually looks like at enterprise scale – in large, multi-account AWS environments with sprawling DNS footprints – and the automation and hard-won triage lessons the problem demands.

Let’s be honest: ~95% of subdomain takeover findings are noise – 3rd- and 4th-level subdomains on dead non-prod environments no customer ever visits. They generate alert fatigue and little else. The other 5% – cookie theft via a *.domain.com scope, OAuth token hijack, phishing from a legitimate domain carrying a valid TLS cert – are catastrophic. Telling the 5% from the 95% is the entire game.

Then there’s the bug bounty dynamic. Researchers find these in bulk and report them in bulk. They’re frustrated that fixes take weeks; program owners are frustrated that most reports are low-impact but can’t simply be closed. The relationship erodes on both sides – a cost nobody puts on a dashboard.

From what We’ve seen, the time sinks are predictable: manually triaging hundreds of researcher reports, chasing low-impact findings, writing one-off fix scripts instead of systemic solutions, and reconstructing record ownership after the fact. The structural causes are just as consistent: multi-account sprawl with no central DNS ownership, rapid dev cycles where IaC teardowns never touch Route53, and researchers reporting faster than teams can remediate.

The core of the talk is architecture: automatically cross-referencing every Route53 A record and CNAME against Elastic IPs allocated across all AWS regions, wired into ownership routing and shift-left pipeline hooks. Done right, this shifts detection from reactive to proactive, can take MTTR from weeks to hours, and eliminates entire classes of records at the source.

We’ll release Ghost Records, an open-source tool aimed at the one of the highest impact vector: dangling A records and CNAMEs pointing to released or unallocated AWS Elastic IPs. It inventories every EIP across all enabled regions, cross-references them against Route53, and flags any DNS record pointing to an IP the account doesn’t own. Read-only, multi-account, parallel scanning, risk-scored output. Live demo included.

People:
    SpeakerBio:  Jai Kumar Sharma, Principal Offensive Security Engineer at GoDaddy

Jai Sharma is a Principal Offensive Security Engineer at GoDaddy, where he leads cloud penetration testing, red team operations, AI red teaming, and security research across one of the world’s largest domain registrars and hosting platforms. A self-taught offensive security researcher from New Delhi, India, he built his career with a hacker’s mindset, breaking code logic, chasing vulnerabilities, and proving real-world business impact from technical exploits. He also leads TTP research, drives threat emulation efforts, and works closely with blue teams to sharpen detection through an adversary-focused lens.

At GoDaddy, Jai tests the same infrastructure and enterprise environments that millions of customers rely on. His work spans cloud and on-premises penetration testing, IAM privilege escalation research, and building automated offensive tooling that helps turn point-in-time assessments into continuous detection. Before GoDaddy, Jai held offensive security roles at Housing.com, FIS, and EY, giving him firsthand experience with how security weaknesses and misconfigurations surface across different industries, architectures, and maturity levels.

Jai volunteers with the DEF CON community as CTF Ops for the Cloud Village and on-site Coordinator for the Bug Bounty Village.

SpeakerBio:  Tom McCarthy

Tom McCarthy is the Director of Offensive Security and Business Information Security Officer at GoDaddy, where he oversees penetration testing, red team operations, and security strategy across multiple lines of business. With over 15 years in offensive security — spanning consulting, penetration testing, research, and training — Tom has built multiple penetration testing teams from the ground up and led or managed hundreds of penetration tests and red team engagements throughout his career. Tom is a returning DEF CON speaker, having previously presented research, tooling, and training at both DEF CON and DerbyCon. He has contributed to open-source offensive security projects including smbexec and Metasploit, and served as a trainer in hacking and incident response for local, federal, and military organizations across multiple countries.

Creator: Nix Vegas Community

Have you ever experienced being rate limited by GitHub? Has GitHub downtime prevented you from building software that you rely on? Are you worried about supply chain attacks? At Exa, we developed a secure, hostable, tarball cache that insulates us from GitHub outages, decreases our build times, protects us from force-pushes upstream, and increases download speeds.

People:
    SpeakerBio:  Ethan Carter Edwards

Ethan is a FLOSS advocate, longtime Nix user and contributor, and engineer. He’s an active member of the Nixpkgs CUDA, Darwin, and NGI teams and is involved in various other efforts throughout the Nix ecosystem.

He currently works on building the infrastructure for the future of websearch at Exa.ai and studies Computer Science at Harvard University.

Creator: ICS Village

Every major ICS attack of the last decade succeeded not because of software vulnerabilities, but because industrial protocols were built to trust any packet on the wire. The village has read the incident reports. What it doesn’t have is a way to replay them against its own infrastructure to learn what its detections actually catch and what they miss.

We present mrhOTshOT, an open-source framework that emulates history’s most destructive ICS attacks across the complete kill chain, reconstructed from publicly available incident analyses. Not just the OT payload the full chain: Windows initial access with real CVEs, lateral movement to engineering workstations, protocol-native process manipulation, and persistent physical impact. Every emulation generates wire traffic consistent with publicly documented behavior on the correct industrial protocol for that attack family.

The framework spans a wide range of industrial protocols across ten distributed PLCs, each simulating the real-world process that protocol actually controls: a heating district controller for Modbus, a safety instrumented system for TriStation, a centrifuge cascade for S7comm. Nothing runs on a generic simulated tank with ten protocols bolted on.

We also introduce the Agentic Attack Emulation Framework: every protocol action is exposed as a callable tool, orchestrated by an LLM agent that reads live process state and composes attack sequences on the fly. No hardcoded playbook, you decide. This is what AI-assisted ICS attack composition looks like, and defenders need to understand it before they meet it in the wild.

The talk closes with a live demo: three centrifuges destroyed in real time while the operator HMI, deceived by an S7 rootkit, shows normal operation throughout, until it doesn’t.

People:
    SpeakerBio:  Malav Vyas, Palo Alto Networks

OT and IoT security researcher. I spend my time hunting vulnerabilities, crafting exploits, and pushing the boundaries of ecosystem security. When I’m not breaking things, I’m helping run the show with the Null, BSides, and OWASP communities. Catch me at the villages.

SpeakerBio:  Asher Davila, Vulnerability Researcher at Palo Alto Networks

Passionate about binary analysis, binary exploitation, reverse engineering, hardware hacking, retro computing, and music.

Creator: Cloud Village

Cloud resources rely on a web of trust that most security teams ignore. Even if you secure access to a resource, you might miss its “shadow dependencies” – the external, system-generated resources required for it to function.

GCP Dataflow is just the latest in a long trail of vulnerable services suffering from this architectural blind spot. What we found is a fundamental flaw in how Dataflow blindly trusts dependencies located far outside its IAM control and security boundaries.

In this talk, we’ll present two novel attack techniques that weaponize unvalidated config files to hijack trusted data pipelines. We’ll also demonstrate a critical cross-tenant vulnerability that extends this issue by enabling manipulation of Dataflow orchestration across tenant boundaries. The issue remains under responsible disclosure and is expected to be fixed before the presentation.

Background: GCP Dataflow and Managed Infrastructure

Organizations running large-scale enterprise data workloads increasingly rely on managed data-processing services, especially GCP Dataflow. Built on Apache Beam, Dataflow executes unified data pipelines while removing much of the operational burden of provisioning cluster frameworks, tuning virtual machines, and scaling infrastructure in response to demand. Teams define the pipeline, and Dataflow provisions and scales the underlying compute resources automatically.

During this orchestration, however, Dataflow pipelines often depend on objects stored in standard cloud storage buckets to control execution, such as code, templates, and environment configuration.

Pipelines routinely ingest and execute files from cloud buckets, including JavaScript or Python User Defined Functions (UDFs) for runtime transformation, structured YAML job templates that define pipeline parameters, and other environment configuration files.

Because these files and their supporting temporary assets are often managed through automated developer pipelines, they can become “shadows”: overlooked in routine asset inventories and security posture reviews, yet still critical to the execution of highly trusted cloud workflows.

Key Takeaways

Dismantling the shadow resources blind spot: Understand how automated cloud orchestration creates short-lived access windows that can evade scheduled security scans and routine asset inventories.

Auditing the infrastructure-as-config attack surface: Learn to treat external execution blueprints as active attack surfaces rather than passive configuration files.

Applying defensive best practices: Take away architectural recommendations for designing pipelines that are more resilient to these attack paths.

Session Details

Total duration: 20 minutes

5 minutes: Background on shadow resources and Dataflow mechanics.

10 minutes: Walkthrough of the attack paths and the cross-tenant vulnerability.

5 minutes: Research methodology, defensive takeaways, and detection strategies.

People:
    SpeakerBio:  Gil Weizman

Gil Weizman is an experienced security researcher with over ten years of expertise across on‑prem, cloud, web, and SaaS security. Gil focuses on threat detection, product security research, vulnerability research and identifying gaps in security visibility across modern environments. Gil specializes in translating real‑world attacker behavior into improved detection and mitigation strategies.

SpeakerBio:  Tamir Yehuda

Tamir Yehuda is a Senior Security Researcher and cloud security team lead at Varonis. He is a full-stack hacker with experience in malware analysis, Windows & AD domains, SaaS applications, and cloud infrastructure. Tamir specialize in IaaS & PaaS research focusing mainly on Azure, GCP, AWS, Salesforce, and ServiceNow. He brings over eight years of experience in various types of security research and red teaming.

Creator: Crypto & Privacy Village

People:
    SpeakerBio:  CPV Gold Bug Team 2026
No BIO available
SpeakerBio:  TBD
No BIO available

Creator: Recon Village

Most LLM-powered recon tools look impressive in a 90-second demo and fall apart on real targets. They hallucinate, loop, go out of scope, double-fire the same endpoint, or die the moment a WAF or rate limit appears. This talk is about what it actually takes to run fifty-plus specialist agents in parallel for hours or days without the chaos.

We built a system where every agent is locked to the rails: a concrete tool, a strict pipeline phase, explicit prerequisites, timeouts, and sandboxes. The model does not drive. It rides. The binaries (and only the binaries) touch the target. This “Agents on Rails” approach eliminates freestyle LLM behavior while still letting the model do what it is good at—reasoning over evidence.

People:
    SpeakerBio:  Anthony Russell, Cyber Security Software Engineer

Anthony Russell, is a senior cyber security engineer with over 13 years of professional experience building software. He has a focus in information security and has been featured in 2600 magazine, on Hak5 and has spoken at both Defcon and DerbyCon multiple times. Favorite things to discuss are blockchain technology, baking custom IoT devices, and everything infosec. You can see more of Anthony’s work at SquidHacker.com or Twitter.com/DotNetRussell

Creator: DCNextGen

Win an AWOK Dual C5 just by showing up! Join Adventures of Illya for Hack the Airwaves! Explore wireless technology with live demos featuring the AWOK Dual C5, ByteShield, H4M PortaPack, Flipper Zero, ESP32 boards, LoRa, RFID, NFC, and CC1101. Learn how these technologies work, ask questions, and get inspired to build your own projects.

People:
    SpeakerBio:  Adventures of Illya
No BIO available

Creator: Radio Frequency Village

What does it take to bounce a signal off Venus and hear it back on Earth? Quite a lot, actually. And we’re going to show all of it to you.

In 2008, AMSAT-DL aimed a carrier wave from a dangerously powerful magnetron-based transmitter, located at the Bochum Observatory, at Venus. Four minutes later, they heard an echo. This was the first time amateurs had bounced a signal of any type off of another planet. This is a significant and enduring achievement.

However, a radar signal has no information in it. It’s a dead carrier. What does it take to bounce a communications signal off of Venus, and hear it back here on Earth? First of all, the path loss is immense. Then, Venus only returns about 13% of the signal that hits it. And, it’s spinning. Information in a signal requires more signal to noise ratio when compared to a dead carrier, nd reflecting off a spinning surface damages information signals in particular ways that stationary reflectors do not.

Venus and Earth encounter each other in their orbits only once every 18 months or so. This is the inferior conjunction of the two planets. Unlike the Moon, frequently used by radio enthusiasts as a reflector, Venus is barely close enough every 18 months for the largest amateur and citizen science dishes to reach. This makes timing, preparation, and coordination equally as important as resolving the technical challenges.

What are the different parts of an Earth-Venus-Earth digital communications system? Which parts make the most difference? What needed to be developed? How did coordination go? What will happen next? Come to the talk and find out how we are going to Hack a Planet.

Quite a lot, actually. And we’re going to show all of it to you.

Links:
    Github – https://github.com/OpenResearchInstitute/EVE

People:
    SpeakerBio:  Abraxas3d, CEO ORI

Michelle Thompson, W5NYV, is the CEO of Open Research Institute, a 501(c)(3) nonprofit developing open source digital radio and space communication systems. ORI’s work directly benefits the amateur radio and amateur radio satellite services. She served as IEEE San Diego Section Chair and represents ORI on the FCC Technological Advisory Council, including co-chairing the AI/ML Safe Uses Sub-Working Group. Her technical background spans RF engineering, digital signal processing, FPGA development, and satellite communications, with prior industry experience at Qualcomm, Smithville Telephone, and Apiary. She can be reached at w5nyv@arrl.net.

Creator: Bug Bounty Village

Building AI-Powered Penetration Testing Bots

In this talk, we’ll walk through the core design philosophy behind AI hackbots and the architecture that makes them work: single-purpose vs. multi-stage bots, context engineering for targeted prompting, and tool integration with output parsing workflows.

Then we’ll get hands-on. We’ll live-build a functional hackbot for a common offensive task —demonstrating asset discovery, endpoint analysis, or mutation-focused testing (e.g., XSS/SSRF) — and show how context engineering and hallucination mitigation work in practice against real targets. You’ll see where AI genuinely accelerates reconnaissance and web analysis, and where it falls flat if you aren’t careful. We’ll close with lessons on cost optimization, auditability, and integrating hackbots into actual engagements.

Who should attend: Pentesters and bug bounty hunters with offensive security experience who want to meaningfully integrate AI into their workflow — not as a novelty, but as reliable tooling.

What you’ll walk away with: A clear mental model for when and how to build hackbots, a live demo you can replicate, and a path into the full course where you’ll build seven production-ready bots from asset discovery through mutation testing.

People:
    SpeakerBio:  Jason “jhaddix” Haddix, CEO and “Hacker in Charge” at Arcanum Information Security

Jason Haddix AKA jhaddix is the CEO and “Hacker in Charge” at Arcanum Information Security. Arcanum is a world class assessment and training company.

Jason has had a distinguished 20-year career in cybersecurity previously serving as CISO of FLARE, CISO of Buddobot, CISO of Ubisoft, Head of Trust/Security/Operations at Bugcrowd, Director of Penetration Testing at HP, and Lead Penetration Tester at Redspin. He has also held positions doing mobile penetration testing, network/infrastructure security assessments, and static analysis. Jason is a hacker, bug hunter and currently ranked 57th all-time on Bugcrowd’s bug bounty leaderboards. Currently, he specializes in recon, web application analysis, and emerging technologies. Jason has also authored many talks on offensive security methodology, including speaking at cons such as DEFCON, Bsides, BlackHat, RSA, OWASP, Nullcon, SANS, IANS, BruCon, Toorcon and many more.

SpeakerBio:  Ryan “BadAt_Computers” Bonner, Lead Security Engineer, Arcanum Information Security

Ryan is a Lead Security Engineer at Arcanum Information Security. A part-time bug hunter, Ryan has built his career on the offensive side of security, focusing on where AI systems and integrations meet the modern web. He specializes in attacking AI-powered applications, including prompt injection, agent, and integration weaknesses that appear as large language models get wired into real products.

Creator: Hackers.town

People:
    SpeakerBio:  RemoteNemesis
No BIO available
SpeakerBio:  Medus4
No BIO available

Creator: Aerospace Village

One of the challenges of independent airplane cyber research is the lack of availability of recent hardware; avionics, LRUs and anything from the aircraft control domain is insanely expensive, even when used. Access to retired airframes therefore represents the state of the art from 20+ years ago. We have been stuck with researching older ACD protocols such as ARINC 429 and 629. However, through a fortuitous stroke of luck, we were given access to an ARINC 664 or ‘AFDX’ environment on a test bench recently. The protocol was developed by Airbus for the A380, but is also found on the B787, A350 and is increasingly being implemented on new designs. Avionics Full-Duplex Switched Ethernet / ADFX will be much more familiar to IT folks than the earlier protocols, having much more in common with the OSI reference model. But, it has crucial differences which requires a steep learning curve. This talk is a primer for interfacing with AFDX and the various security and safety features that it offers.

People:
    SpeakerBio:  Andrew Tierney
No BIO available
SpeakerBio:  Adam Bromiley, Pen Test Partners

Adam is a security consultant at Pen Test Partners who specialises in industrial control systems and embedded hardware security. He's worked on everything safety-critical: from high-speed rail to aircraft, power stations, and gas distribution. His embedded work has seen him break driverless cars, slot machines, and drones and has led to the responsible

disclosure of numerous vulnerabilities in industrial controllers. Adam enjoys hands-on and boots-on-the-ground testing, reverse engineering, and providing practical security advice for complex real-world systems.

Creator: Aerospace Village

NTSB Vehicle Recorder Specialists Jonathan Xue and David Case will give a talk on decoding data recovered from several aircraft accidents, and a tourist submarine raised from the depths of the North Atlantic. The NTSB’s process for getting data out of broken and damaged equipment will be shown, including searching through gigabytes of random data to find a log, and then converting that data to something human readable. Jonathan and David specialize in converting undocumented binary data recovered from accidents into graphs and charts usable in investigation. Quite often, they are the first people to see what actually happened in an accident.

People:
    SpeakerBio:  David Case, NTSB
No BIO available
SpeakerBio:  Jonathan Xue, NTSB
No BIO available

Creator: .EDU Community

People:
    SpeakerBio:  Dave “Rebelcadet” Ortiz

David Ortiz is the Chief Technology Officer (CTO) at X8 LLC, where he leads the firm’s engineering efforts to develop innovative technical, operational, and research solutions to meet complex customer requirements. With a diverse background in laser physics, information assurance, cyberspace operations, and organizational leadership, David is a seasoned professional dedicated to advancing cyber capabilities for both the U.S. Government and private industry customers.

Creator: Aerospace Village

Commercial aircraft have Traffic Collision and Avoidance Systems, or TCAS, to ensure they don’t fly in to each other, but these are very expensive systems. How do General Aviation aircraft find a way to avoid each other?

Enter ‘Electronic Conspicuity’ devices: small, light and cost effective sensors to detect other aircraft and alert GA pilots before they can be seen by the naked eye.

Unfortunately, in the rush to improve flight safety and situational awareness in GA, cyber security has taken something of a back seat. As a result, some of these EC devices have trivial and not-so-trivial security flaws. In some cases, it is possible to delete Traffic Advisories and/or give an incorrect Resolution Advisory, bringing planes in to conflict with each other, rather than ensuring they stay safely apart.

People:
    SpeakerBio:  Ken Munroe
No BIO available

Creator: Biohacking Village

Gradually we are all becoming more and more dependent on connected technology. We will be able to live longer with an increased quality of life due to medical devices and sensors attached to, or integrated into our bodies. However, our dependence on technology grows faster than our ability to secure it, and a security failure of a medical device may cause patient harm and have fatal consequences.

This presentation dives into the security architecture of the Biotronik pacemaker ecosystem, covering the pacemaker, home monitoring units, and external programmer. Using a hybrid of black-box and white-box methodologies, we deconstruct the firmware and wireless communication protocols to identify vulnerabilities in a pacemaker that one of the presenters was depending on with their life for 11 years. We will discuss the challenges of extracting firmware from life-critical hardware and the implications of discovered bugs on patient safety. Attendees will leave with a better understanding of how to reverse engineer proprietary medical ecosystems and why securing the “Internet of Medical Things” is a race we cannot afford to lose.

People:
    SpeakerBio:  Marie Moe

Dr. Marie Moe is a Principal Consultant at Mandiant (now part of Google Cloud), and has a PhD in information security. She is a part-time Associate Professor at NTNU. In this talk she will be joined by NTNU MSc students Shayan Alinejad and Kristian Karlsen.

SpeakerBio:  Shayan Alinejad
No BIO available
SpeakerBio:  Kristian Karlsen
No BIO available

Creator: Bug Bounty Village
Customer support has always been a numbers game: high message volume, repetitive questions, and constant pressure to respond faster. Most organisations handle this through ticket portals, shared mailboxes or live chat tools. These systems were built around people reading messages and people taking action. That model is disappearing. To keep up with demand, support platforms now embed AI agents that read incoming messages, suggest replies, search internal knowledge pages and even interact with backend systems. They reset accounts, schedule refunds, modify settings and verify users before a human ever looks at the conversation. The human is still in the loop, but rarely in full control. Connecting a language model to business logic has a side effect. It turns text into a control layer. Messages no longer just inform a conversation; they influence actions. If an attacker can shape those messages, they can shape what the agent does. In security terms, that creates a new attack surface that did not exist when humans handled every request. This session examines how that surface is exploited in practice. The findings come from real vulnerability reports against production AI support systems that accept contact from the public. The attacks require no insider access. They target the assumptions that connect users, agents and operators, and they convert helpful automation into unintended system access. The examples that follow show how the trust built into these systems can be bent until it breaks.

People:
    SpeakerBio:  Inti “securinti” De Ceukelaire, Founding Member, Intigriti

Intigriti is a Founding Member at Intigriti and their former Chief Hacker Officer. As a hacker, Inti has won multiple awards in live hacking competitions from Bugcrowd and HackerOne. Inti specialises in AI-hacking techniques, logic flaws and support systems, and is most known for the “Ticket Trick” methodology that has earned a spot in Portswigger Top Web Hacking Techniques.

Creator: Red Team Village

AI defenses are only as good as the analysts training them. This session exposes how attackers manipulate SOC analyst behavior to silently degrade AI baselines over time, no exploit required. We cover the mechanics of false positive injection, alert fatigue as a weapon, and misdirection. Then attendees step into the analyst’s seat in a live browser-based simulation and experience how their own decisions poison the model.

People:
    SpeakerBio:  Alexander “Zombie”

4.5 years in the SOC. Junior Analyst to Team Lead. CPTS, CRTO, CRTE, OSCP and enough certs to wallpaper a small room with some left over. Zombie has spent the better part of half a decade watching how defenders think, how they triage, how they get tired, and how attackers can weaponize all three. By day he leads a SOC team. By night he’s learning everything he can about breaking the things he built. Still technically blue team. Emphasis on technically.

SpeakerBio:  Lee McWhorter

Lee McWhorter, Owner & Chief Geek at McWhorter Technologies, has been involved in IT since his early days and has over 30 years of experience. He is a highly sought after professional who first learned about identifying weaknesses in computer networks, systems, and software when Internet access was achieved using a modem. Lee holds an MBA and more than 20 industry certifications in such areas as System Admin, Networking, Programming, Linux, IoT, and Cybersecurity. His roles have ranged from the server room to the board room, and he has taught for numerous universities, commercial trainers, and nonprofits. Lee is the SWAG Master and a Core Staff member for the Red Team Village at DEFCON; and works closely with the Pacific Hackers Association, Dark Arts Village at RSAC, Texas Cyber Summit, CompTIA, and the CompTIA Instructor Network as a Speaker, SME, and Instructor.

Creator: Red Team Village

In this talk, we introduce Haetae, an agent designed to profile, identify, and exploit C2 frameworks used by North Korean malware.

Haetae supports both automated and interactive modes, allowing analysts to map and understand adversary infrastructure with different levels of control. It is built around a flexible rule-based system, enabling users to extend and refine detections as new patterns and frameworks emerge.

In this first release, we will walk through real-world cases where Haetae was used to identify and take down infrastructure associated with Mach-O Man and POWerful Armadillo.

We will also release a safe emulator of both malware C2 servers, allowing researchers and newcomers to experiment with Haetae in realistic, controlled environments without risk.

This is a highly technical talk but can be enjoyed by both beginners and seasoned threat hunters.

People:
    SpeakerBio:  Mauro Eldritch, Leader at Bitso Quetzal Team

Hacker and Speaker.

Founder of BCA LTD and DC5411.

I wrote a book interviewing Threat Actors.

I like Threat Intelligence and Golden Retrievers.

SpeakerBio:  Nelson Rafael Colón Merán, Security Engineer at Bitso

Security Engineer at Bitso, Latin America’s leading crypto-first financial platform. I’ve specialized in red team operations, penetration testing, and malware development.

Recognized speaker in the Dominican Republic’s RedTeamRD cybersecurity community, where I’ve given a couple talks and previously assisted DEFCON as a speaker.

Creator: Adversary Village

In this talk, we introduce Haetae, an agent designed to profile, identify, and exploit C2 frameworks used by North Korean malware.

Haetae supports both automated and interactive modes, allowing analysts to map and understand adversary infrastructure with different levels of control. It is built around a flexible rule-based system, enabling users to extend and refine detections as new patterns and frameworks emerge.

In this first release, we will walk through real-world cases where Haetae was used to identify and take down infrastructure associated with Mach-O Man and POWerful Armadillo.

We will also release a safe emulator of both malware C2 servers, allowing researchers and newcomers to experiment with Haetae in realistic, controlled environments without risk.

This is a highly technical talk but can be enjoyed by both beginners and seasoned threat hunters.

People:
    SpeakerBio:  Mauro Eldritch, Leader at Bitso Quetzal Team

Hacker and Speaker.

Founder of BCA LTD and DC5411.

I wrote a book interviewing Threat Actors.

I like Threat Intelligence and Golden Retrievers.

SpeakerBio:  Nelson Rafael Colón Merán, Security Engineer at Bitso

Security Engineer at Bitso, Latin America’s leading crypto-first financial platform. I’ve specialized in red team operations, penetration testing, and malware development.

Recognized speaker in the Dominican Republic’s RedTeamRD cybersecurity community, where I’ve given a couple talks and previously assisted DEFCON as a speaker.

Creator: Malware Village

In this talk, we introduce Haetae, an agent designed to profile, identify, and exploit C2 frameworks used by North Korean malware.

Haetae supports both automated and interactive modes, allowing analysts to map and understand adversary infrastructure with different levels of control. It is built around a flexible rule-based system, enabling users to extend and refine detections as new patterns and frameworks emerge.

In this first release, we will walk through real-world cases where Haetae was used to identify and take down infrastructure associated with Mach-O Man and POWerful Armadillo.

We will also release a safe emulator of both malware C2 servers, allowing researchers and newcomers to experiment with Haetae in realistic, controlled environments without risk.

This is a highly technical talk but can be enjoyed by both beginners and seasoned threat hunters.

People:
    SpeakerBio:  Nelson Colon, Offensive Security Specialist at Bitso Quetzal Team

Dominican Offensive Security Specialist. I spoke at DEF CON Data Duplication Village about creating immortal C2 servers using modern data duplication platforms.

Creator: Ham Radio Village

The most common question we hear in the village is, “What can I do now that I’m licensed?” N0OPS and W0BDP attempt the guide new hams into incorporating ham radio into their current hobbies or pursuing new interests.

N0OPS and W0BDP will walk the audience through the diversity of ham radio operations such as portable activations such as POTA/SOTA, satellite contacts, digital weak-signal modes, direction-finding competitions, emergency communications, and drone telemetry.

The takeaway: no matter how long you’ve held a call sign whether it’s 5-minutes or 5 years, there’s almost certainly a corner of this hobby you haven’t touched yet. This speech is an invitation to go find it.

People:
    SpeakerBio:  Dan W0BDP Norte

Dan “darkestofdans” Norte (W0BDP) is a penetration tester, currently with NetSPI. Dan has been a ham radio operator for 9 years, currently Amateur Extra class. Dan enjoys popping shells and chasing DX.

Creator: Ham Radio Village

This talk approaches off-grid communications as a practical systems problem, informed by the author’s real-world operation of portable and off-grid radio systems rather than theory or gear marketing. Rather than teaching amateur radio as a hobby, it reframes off-grid communications using failure-mode analysis and systems design principles familiar to the security and hacking community. While the question “How do I talk 500 miles when the grid is down?” is a common starting point, the deeper issue is how communication systems behave when infrastructure is absent, degraded, or intentionally avoided — and why many well-intentioned plans fail in practice.

The presentation is structured as a 101-level introduction for a technically curious audience. It assumes no prior amateur radio experience, but it does assume interest in understanding constraints, tradeoffs, and failure modes the same way hackers evaluate any other system operating outside ideal conditions.

The talk opens by examining real examples from off-grid and preparedness communities, including common questions and the unhelpful responses they often attract. This framing establishes the core thesis: both “prepper fantasy” and “sad ham” gatekeeping optimize for the wrong constraints and lead to brittle designs that fail when conditions are imperfect.

From there, the talk builds a mental model of off-grid radio communications grounded in physics and operational reality. Attendees are introduced to why radio performance is probabilistic rather than guaranteed, why a fixed distance like “500 miles” is a misleading requirement, and how factors such as propagation, noise floor, antenna efficiency, duty cycle, and coordination dominate outcomes more than transmitter power or radio model. Real failure modes are discussed, including HF voice links failing when expected to work, digital modes succeeding where voice does not, antenna quality overwhelming other variables, high-noise environments negating otherwise capable setups, and situations where increasing power does not meaningfully improve reliability.

The scope then expands beyond emergencies to everyday off-grid use cases hackers actually encounter: group communications while hiking or camping without cell coverage, portable and battery-powered operation, short-range mesh and store-and-forward systems such as Meshtastic, and longer-range low-bandwidth HF messaging. Activities like Parks on the Air, Summits on the Air, and Field Day are presented not as hobbies, but as structured opportunities for exploration and experimentation — real-world environments for testing antennas, power systems, logistics, and coordination under off-grid constraints.

Rather than focusing on specific products or modes, the talk emphasizes design principles that consistently work: simple plans over clever ones, redundancy across bands and modes, asynchronous messaging over real-time voice, and communication plans that account for human behavior as a primary constraint. Regulatory considerations are covered at a practical level to help attendees understand what is possible and appropriate without turning the talk into a licensing lecture.

If appropriate for the venue, the talk may include a short demonstration or case study illustrating propagation variability or the effectiveness of low-bandwidth digital techniques compared to voice. Any demonstration will be designed with realistic fallback options and framed as illustrative rather than guaranteed.

The goal of this talk is not to turn attendees into radio experts, but to give them a reusable framework for thinking about off-grid communications the way hackers think about resilient systems: define the actual problem, understand how and why things fail, and design solutions that work in imperfect, real-world environments.

Every hacker has seen the post: someone asks what radio will let them “reliably talk 500 miles to family when the cell network goes down.” The replies quickly devolve into bad advice, magical thinking, and dismissive gatekeeping. The result is confusion, wasted money, and false confidence about off-grid communications.

This talk is the antidote.

We’ll cut through the mythology and explain what amateur radio can actually do for personal and family communication during major disruptions — and what it cannot. Starting at a 101 level, we’ll cover realistic ranges, basic propagation, equipment tradeoffs, digital modes, licensing myths, and why distance is usually the wrong requirement.

Rather than dunking on preppers or policing fun like a sad ham, this talk reframes the problem the way hackers do: understanding constraints, failure modes, and human factors.

We’ll focus on practical off-grid communication strategies that work in the real world — coordination, redundancy, and low-bandwidth messaging — not fantasy continent-spanning voice links.

People:
    SpeakerBio:  Andrew Ohnstad – N3OCQ

Andrew has spent 25 years designing large-scale IT systems. He grew up alongside the birth of the internet, building his own computers, running a BBS, and taking electronics apart to understand how they worked. His interests remain firmly hands-on, focused on hardware tinkering and RF. He’s been a licensed amateur radio operator since 1993, holds an Amateur Extra license, and is active from VHF/UHF through 160 meters.

Creator: Red Team Village

Pentest Copilot is an MIT licensed open-source agentic penetration testing workspace designed to assist security professionals with web/network application assessments.

This hands-on tactic introduces participants to operating autonomous web pentesting workflows against intentionally vulnerable applications. Attendees will deploy Pentest Copilot, configure the environment, and use it to perform attack-surface discovery, authenticated testing, vulnerability investigation, and workflow-driven analysis.

Rather than focusing on theory, the session is designed around practical offensive workflows. Participants will observe how autonomous agents navigate modern web applications, identify interesting attack paths, investigate findings, and assist with security testing tasks that traditionally require significant manual effort.

The goal is to provide attendees with a realistic understanding of how autonomous systems can augment offensive security work today, where they are effective, and where human operators remain essential.

Pentest Copilot is fully open source and available at: https://github.com/bugbasesecurity/pentest-copilot, Wiki is at: https://github.com/bugbasesecurity/pentest-copilot/wiki over 800+ stars.

People:
    SpeakerBio:  Dhruva Goyal

Dhruva has been hacking since middle school and enjoys offensive security research and red teaming. Dhruva is certified with OSCE3(ie. OSWE, OSEP & OSED) and OSCP. He leads the cybersecurity engagements and triage team at BugBase. He also loves playing AOE4 and working out.

SpeakerBio:  Sitaraman Subramanian

Sitaraman Subramanian is the CTO at BugBase. He has spent over five years working in offensive security and over a decade building products. His work spans full stack development, DevSecOps, cloud security, and offensive security. He built BugBase from the ground up, and now leads engineering on Pentest Copilot, an autonomous AI driven pentesting platform. He presented at Black Hat USA Arsenal in 2025 and Microsoft BlueHat in 2024. He is an active open source contributor. Pentest Copilot is open sourced at https://github.com/bugbasesecurity/pentest-copilot, with 850+ stars and growing. He writes about hacking and AI evals at https://blog.ssitaraman.com.