BEGIN:VCALENDAR
VERSION:2.0
PRODID:Data::ICal 0.24
BEGIN:VEVENT
DESCRIPTION:   'Title: Total Recon: How We Discovered 1000s of open Agents 
 in The\n   Wild\n   Tags: Recon Village | Creator Talk/Panel\n   When: Sat
 urday\, Aug 8\, 17:30 - 17:59 PDT\n   Where: LVCCW Level 1 Hall 3 1105 (Cr
 eator Stage 3) - [1]Map\n\n   Description:\n   AI agents quietly created a
  new external attack surface: copilots\,\n   custom agents\, AI cakends an
 d various deployments that ship to the\n   internet\, often without anyone
  realizing they are reachable\,\n   enumerable\, or over-permissive.\n\n  
  In this talk\, we’ll show how attackers can already find your agents\n 
   in the wild\, shedding light on the technical details that enable this\n
    kind of malicious activity\, including how we used these details to\n  
  find 1000s of exposed agents of different kinds. We’ll follow up\n   wi
 th explaining how to measure exposure\, see the proof for obscurity\n   fa
 iling\, and understand how to detect threat-actor agent-focused recon\n   
 before it turns into an impactful attack. Capping it all off by\n   showca
 sing PowerPwn\, a recon tool you can use to test your own\n   exposure\n\n
    Speakers:Avishai Efrat\,Roey Ben Chaim\n\n   SpeakerBio:  Avishai Efrat
 \, Senior Security Researcher at Zenity\n\n   Avishai Efrat is a senior se
 curity researcher at Zenity\, specializing\n   in securing AI agents and l
 ow-code/no-code platforms. His work focuses\n   on uncovering vulnerabilit
 ies in enterprise AI deployments\, from\n   Copilot Studio to ChatGPT and 
 beyond - and exploring how attackers\n   discover and exploit these weakne
 sses. Passionate about all aspects of\n   security\, with experience spann
 ing web security\, anti-bot protection\,\n   OSINT\, blockchain\, and data
  engineering and aim on bridging the gap\n   between cutting-edge research
  and practical defense strategies.\n   Previously presented at BSidesTLV\,
  Black Hat USA & Europe Arsenal and\n   SecTor.\n\n   SpeakerBio:  Roey Be
 n Chaim\, Staff Engineer at Zenity\n\n   Roey Ben Chaim is an engineer foc
 used on AI and agent security\, with a\n   particular interest in how agen
 tic systems fail in practice. His work\n   centers on stress-testing agent
 s\, building benchmarks for long-horizon\n   behavior\, and developing pra
 ctical defenses against risks such as\n   prompt injection\, unsafe tool u
 se\, privacy leakage\, and adversarial\n   workflows. Prior to this\, he s
 pent a decade at Microsoft building\n   large-scale security systems. He i
 s a co-author of SkillsBench\, a\n   benchmark on agent skills efficacy ov
 er diverse tasks\, and contributes\n   to safety and privacy tooling\, inc
 luding Presidio. Roey is also a\n   speaker on agentic systems and AI secu
 rity\, a co-organizer of AI\n   Tinkerers Tel Aviv\, and an active partici
 pant in hackathons and\n   builder communities\, where he explores emergin
 g systems through\n   hands-on experimentation.\n\n   '\n\n   1. #LVCCW_Le
 vel1_Hall3\n\n\n
DTEND:20260809T005900Z
DTSTART:20260809T003000Z
LOCATION:Recon Village - LVCCW Level 1 Hall 3 1105 (Creator Stage 3)
SUMMARY:Total Recon: How We Discovered 1000s of open Agents in The Wild
END:VEVENT
END:VCALENDAR