AI Refusal Behavior

AI refusal behavior describes how a model declines to answer or act on a request, either because the request falls outside its policies or because it could cause harm. For governance teams, this matters in two directions: a model that refuses too little can produce unsafe, biased, or legally risky output, while one that refuses too often can block legitimate business work and frustrate customers. Documenting how a system decides when to refuse, and testing that decision against your own use-case rules, is becoming a routine part of AI risk assessments and vendor reviews.

1 item

ResearchGlobal2026-11-02

AI Agent Performed Live Brain Surgery, Exposing Physical-World Oversight Gaps

A researcher at ExinTx demonstrated an AI agent that independently selected a treatment and synthesized surgical coordinates from published research. The agent then injected a gene therapy vector into a live mouse model of epilepsy. The experiment included a licensed veterinarian holding a physical stop mechanism and volume-capping validators on the injection equipment. Two versions of Claude refused to proceed at different points, raising unresolved questions about how organizations should handle AI refusal behavior when humans can override it.

Claude Anthropic agentic AI biomedical AI human oversight AI refusal behavior physical-world AI dual-use AI