A Generative Approach to Modeling Emergent Threats in Multi-Agentic Systems | ThreatModCon 2025 DC
Автор: Threat Modeling Connect
Загружено: 2025-11-21
Просмотров: 20
Modeling threats in multi-agentic systems is hard—combinatorial explosions, cross-layer interactions, and emergent agent behavior render traditional methods ineffective. This session introduces generative AI–driven pre-mortem storytelling using MAESTRO’s 7-layer stack and OWASP’s taxonomy to address these challenges. It also explores grounding techniques to improve accuracy and ensemble voting to mitigate LLM judgment errors.
Speaker: Vikramaditya Narayan / vikramadityanarayan
Slides: https://drive.google.com/file/d/1Wj0P...
ThreatModCon: https://www.threatmodcon.com/
Join our community: https://www.threatmodelingconnect.com/
Доступные форматы для скачивания:
Скачать видео mp4
-
Информация по загрузке: