2026
PropensityBench: Evaluating Latent Safety Risks in Large Language Models via an Agentic Approach
ICLR 2026poster
Recent advances in Large Language Models (LLMs) have sparked concerns over their potential to acquire and misuse dangerous capabilities, posing frontier risks to society. Current safety evaluations primarily test for what a model *can* do---its capabilities---without assessing what it *would* do if…