Skip to content

Evaluation

Computer use malicious use evaluation

compliance with requests to perform harmful actions (deceptive/fraudulent activity, malware distribution, targeting/profiling, malicious content delivery) via computer use

Agentic & computer use Anthropic

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 3.7 Sonnet
Anthropic

rich detail 16 method fields published

See how this test was run