BEGIN:VCALENDAR
VERSION:2.0
PRODID:Data::ICal 0.24
BEGIN:VEVENT
DESCRIPTION:   'Title: Automated Discovery of Prompt Injection Vulnerabilit
 ies via\n   Mutated Prompt Generation\n   Tags: Bug Bounty Village | Creat
 or Talk/Panel\n   When: Sunday\, Aug 9\, 11:30 - 11:59 PDT\n   Where: LVCC
 W Level 1 Hall 3 1102 (Creator Stage 6) - [1]Map\n\n   Description:\n\n   
 Finding one prompt injection isn't the end goal\; checking how many\n   va
 riants still bypass your safeguarding layers is the new target.\n   Introd
 ucing MPG\, an XPIA mutation framework that generates adversarial\n   payl
 oad variants and tests them across agentic AI systems to expose\n   hidden
  vulnerability surfaces and emerging data exfiltration risks.\n\n   Speake
 rs:Bogdan Stelee\,Arnav Garg\n\n   SpeakerBio:  Bogdan Stelee\, Senior Sof
 tware Engineer\, Microsoft\n\n   Bogdan Stelea is a Senior Software Engine
 er performing AI Security\n   Research at Microsoft\, working at the inter
 section of AI security\,\n   adversarial ML\, and secure agentic system de
 sign. Bogdan develops\n   automated frameworks for conducting variant hunt
 ing and discovering\n   prompt injection vulnerabilities at scale. His wor
 k focuses on\n   identifying\, reproducing\, and understanding vulnerabili
 ties in modern\n   AI copilot solutions\, with a particular emphasis on cr
 oss/indirect\n   prompt injection attacks (XPIA)\, tool misuse\, and the e
 xpanding attack\n   surface introduced by AI agents. Moreover\, Bogdan cre
 ates learning\n   materials and courses that clearly explain the threats i
 ntroduced by\n   XPIA. Actively engaged in the AI Security\, CTF\, and Bug
  Bounty\n   communities\, with a consistent presence at DEF CON\, Bogdan h
 as\n   published and presented his AI security research at international\n
    academic venues\, including IEEE and ACM.\n\n   SpeakerBio:  Arnav Garg
 \, Security Researcher\, Microsoft\n\n   Arnav Garg is a Security Research
 er at the Microsoft Security Response\n   Center (MSRC)\, specializing in 
 AI security research\, adversarial\n   testing\, and the security of Copil
 ot and agentic AI systems. His work\n   focuses on understanding and mitig
 ating emerging AI threats\, with\n   particular emphasis on cross/indirect
  prompt injection attacks (XPIA)\,\n   data exfiltration techniques\, tool
  misuse\, and the evolving attack\n   surface introduced by AI agents. Arn
 av develops automated frameworks\n   for scalable security testing and vul
 nerability discovery\, helping\n   advance practical defenses for next-gen
 eration AI systems. He has\n   presented and published research on AI secu
 rity\, automated prompt\n   injection discovery\, and AI vulnerability tes
 ting at both internal and\n   external venues.\n\n   '\n\n   1. #LVCCW_Lev
 el1_Hall3\n\n\n
DTEND:20260809T185900Z
DTSTART:20260809T183000Z
LOCATION:Bug Bounty Village - LVCCW Level 1 Hall 3 1102 (Creator Stage 6)
SUMMARY:Automated Discovery of Prompt Injection Vulnerabilities via Mutated
  Prompt Generation
END:VEVENT
END:VCALENDAR