UROP · CA presentation

Complier: Provable Safety for Interacting AI Agents

Kavish Sathia

Supervisor

Prof Martin Henz

Main evaluator

Prof Michael D Adams