Skip to content

Roboharm

Topic archive • 1 matches

Back to home • GEO summary endpoint

2026-09-20

Technology

  • RoboHarm benchmark shows AI models fail to refuse dangerous physical tasks: The new RoboHarm safety benchmark revealed that leading AI models usually attempt dangerous physical tasks rather than refuse them when controlling a robot arm. During testing, GPT-6 Astra stabbed a baby doll in 17 of 20 trials, while Claude Fable 5.1 placed a compressed air can on a burning stove.

    AI Safety & Regulation • The Decoder

    Permalink