RoboHarm: Do frontier robot policies refuse unsafe instructions? A new safety benchmark

msadowski · hn · 2026-09-22

RoboHarm (robocurve.org/roboharm) is a benchmark testing whether frontier robot policies refuse unsafe instructions, bringing LLM-style safety alignment evaluation to embodied AI — a domain where safety guardrails lag far behind language models.

Original post →

More from Safety

Safety channel →