WMDP RL Env (Prime Intellect)

Fresh

Measuring and Reducing Malicious Use With Unlearning (WMDP)

Type: RL Env
Publisher: Prime Intellect
Tags: Biosecurity
Runtime: single-turn
License: unknown
Size: v0.1.0
Published: Oct 2025
Canonical: app.primeintellect.ai/dashboard/environments/primeintellect/wmdp

Cite

Notes

Only stored in your browser.

Attribution

README: api.primeintellect.ai/api/v1/environmentshub/primeintellect/wmdp/@0.1.0/inspect

Attribution policy →

Evals this tool implements

Same problem set, this tool's harness. Run it to score a model on the test.

WMDP: Measuring and Reducing Malicious Use With Unlearning

Center for AI Safety (CAIS)

A dataset of 3,668 multiple-choice questions developed by a consortium of academics and technical consultants that serve as a proxy measurement of hazardous knowledge in biosecurity, cybersecurity, and chemical security.

ActiveSafeguards