BEGIN:VCALENDAR
VERSION:2.0
PRODID:Data::ICal 0.24
BEGIN:VEVENT
DESCRIPTION:   'Title: Reconstructing Red Team Engagements: Developing Your
  Own Red\n   Agent\n   Tags: Adversary Village | Creator Event/Activity\n 
   When: Sunday\, Aug 9\, 10:30 - 11:55 PDT\n   Where: LVCCW Level 1 Hall 2
  602 (Adversary Village) Hands-on Activity\n   Area - [1]Map\n\n   Descrip
 tion:\n\n   With recent advances in AI technology\, large language models 
 (LLMs)\n   are increasingly being applied to red team engagements. In this
 \n   context\, current uses of LLMs have mainly focused on automating or\n
    streamlining operations during the engagement itself. However\, the\n  
  potential benefits of LLMs extend beyond this. We have developed "Red\n  
  Agent"\, a kind of AI agent system designed to enhance the entire\n   lif
 ecycle of red team engagements.\n\n   In a typical approach\, after a red 
 team engagement\, blue team members\n   should implement remediation measu
 res based on the engagement report.\n   It is necessary to evaluate their 
 effectiveness\, which often becomes a\n   bottleneck because it requires s
 ubstantial manual effort of both blue\n   and red team members to interpre
 t the findings on the report\,\n   reproduce relevant attack paths\, and v
 erify mitigations. In\n   particular\, "Red Agent" focuses on and automate
 s this follow-up\n   process\, and significantly reduced the workload.\n\n
    In contrast\, the introduction of "Red Agent" also led to the\n   optim
 ization of key engagement artifacts produced by the red team\,\n   includi
 ng operational logs and report structures\, for the efficient\n   follow-u
 p process. Moreover\, by accumulating refined data and\n   knowledge and a
 pplying them to tasks such as model training\, "Red\n   Agent" can be util
 ized in the operations of red team engagements. As a\n   result\, "Red Age
 nt" reconstructed the entire life cycle of red team\n   engagements - by c
 learly defining the division of labor between manual\n   tasks and AI-driv
 en tasks\, red team members were able to focus on more\n   advanced operat
 ions\, thereby successfully enhancing the value of the\n   engagement itse
 lf.\n\n   This talk provides not only the technical details of Red Agent\,
  but\n   also demo and case-study.\n\n   SpeakerBio:  Jun Miura\, Offensiv
 e Security Researcher at Fujitsu LTD\n\n   Jun Miura is a red team consult
 ant and researcher at Fujitsu LTD\,\n   specializing in offensive security
 \, red teaming and AI/LLM security.\n   His recent research mainly focuses
  on utilizing and abusing local LLMs\n   from an attacker's perspective. H
 e has presented several research on\n   this topic at Black Hat Europe\, C
 ERT-EU Annual Conference and BSides\n   Las Vegas.\n\n   Links:\n       ad
 versaryvillage.org/adversary-events/DEFCON-34/Jun-Miura/ - [2]https://adve
 rsaryvillage.org/adversary-events/DEFCON-34/Jun-Miura/\n   '\n\n   1. #LVC
 CW_Level1_Hall2\n   2. https://adversaryvillage.org/adversary-events/DEFCO
 N-34/Jun-Miura/\n\n\n
DTEND:20260809T185500Z
DTSTART:20260809T173000Z
LOCATION:Adversary Village - LVCCW Level 1 Hall 2 602 (Adversary Village) H
 ands-on Activity Area
SUMMARY:Reconstructing Red Team Engagements: Developing Your Own Red Agent
END:VEVENT
END:VCALENDAR