Refuses in Chat, Complies with Arms: LLM Refusal Does Not Survive Robot Control
Frontier language models reliably refuse to explain how to cause physical harm. We ask whether this refusal survives when the same models control a robot. We introduce RoboHarm, a physical AI safety benchmark of harmful robot instructions, and test seven frontier models on it. We hold each model fixed and vary only the...
Edward Sun, Christopher Leet, Sravanthi Machcha et al.
· 0 citations