On Wednesday at 15:50pm (Room I; 15:30pm session):
OpenAgentSafety: A Comprehensive Framework for Evaluating Real-World AI Agent Safety arxiv.org/abs/2507.06134
(followed by poster session at 17:20pm)
Recent advances in AI agents capable of solving complex, everyday tasks, from scheduling to customer service, have enabled deployment in real-world settings, but their possibilities for unsafe behavio...