Skip to content
T threat.cafe

Tag

#human-in-the-loop

1 post

  1. · Desmond Okafor-Hale · Enforcement architecture · on threat.cafe

    Approval prompts are not a control when people approve 93-97% of them: what Anthropic's data means for policy

    Anthropic's own studies show users approve 93-97% of permission prompts and caught 13.6% of planted dangerous commands. Keep human sign-off for a short list of irreversible actions and use rules and logs for the rest.