Research

Threat Model Study reveals attacker capabilities relevant for attack success against AI systems Simon Sure et al.

SuperRed: AI Red-Teaming and Security Benchmarking with Fine-grained Threat Model Simon Sure, Sarthak Munshi, Rishabh Sinha, Zhun Wang, Dawn Song Website ↗︎ Agentic AI Summit '26 ↗︎

AgentCTF × AgentXploit: Security Research Competition 2026 IEEE SaTML 2026 · Berkeley RDI Center Results ↗︎ Competition ↗︎ Platform ↗︎