Rohan Paul (@rohanpaul_ai) · 10/16/2026, 02:38:53
Anthropic Grants METR Evaluators Internal Access and Unblockable Public Verdict Rights
Anthropic has embedded six safety evaluators from the nonprofit METR inside its company, granting them internal access and resources comparable to its own risk teams. METR retains the right to publish independent verdicts on model risks, which Anthropic cannot block but can only redact within predefined categories. This move signifies a shift in AI third-party auditing from black-box testing to white-box residency.
SOURCE COVERAGEFull Original Report
Rohan Paul (@rohanpaul_ai) · 2026-10-15 18
Anthropic has placed 6 safety evaluators from the outside nonprofit METR inside the company, with access to internal systems, tools and staff roughly equal to that of its own risk teams.
Six METR staff have company hardware, office space, and internal access “roughly on par” with Anthropic’s own risk assessment teams.
METR will publish its own verdicts on how risky Anthropic’s models and practices are, and Anthropic cannot block them, only redact a few predefined categories such as customer data, a decision METR can publicly challenge.
Other evaluators, such as SecureBio, start at the lowest of 3 new access tiers, where Anthropic still controls what they see and must approve what they publish.
↳ 引用 @AnthropicAI (Anthropic): We’ve committed to having evaluators assess our safety practices. We're in discussions with multiple organizations, and have now given employee-like access to a team from @METR_Evals.
In this post, we describe how we work with third-party evaluators: https://www.anthropic.com/news/an-update-on-independent-evaluators
