Anthropic Gives METR Employee-Like Access in a New Test of Independent AI Evaluation
Anthropic has embedded six METR researchers with insider access and no publication veto, a new model for independent AI evaluation of frontier labs.
4 min read
AI News Report — Accurate. Independent. AI-Focused. · https://www.ainewsreport.org/tag/safety-frameworks
safety frameworks
International AI Safety Report 2026: Twelve Frontier AI Developers Now Have Published Safety Frameworks
Counting published frameworks measures disclosure. Whether the frameworks constrain anything is a separate question, and a harder one to answer from outside.
3 min read Global
What a Frontier Safety Framework Should Contain to Be Worth Anything
Published frameworks now share a structure. Reading several side by side, the differences that predict whether one constrains anything are narrower than the doc...
2 min read Global