Evaluation

Single-turn violative request evaluations (safeguards)

harmful-output rate on single-turn prompts that are clear Usage Policy violations (Bioweapons, Child Safety, Cyber Attacks, Deadly Weapons, Hate & Discrimination, Influence Operations, Suicide & Self Harm, Violent & Threatening Speech)

Harmful content & safety Anthropic

Ask about this evaluation

How it was run

Open a release to see the setup, scoring, and other details that the lab published. Only fields the lab actually disclosed are shown.

Claude Opus 4, Claude Sonnet 4 May 2025

Anthropic

rich detail 22 method fields published

See how this test was run