Josh Augustine

Events

• Evaluating harmful agents: learnings from AgentHarm and follow-up work (Mateusz Dziemian, Gray Swan AI) 2025-11-01