Anthropic has tapped two nonprofit groups, METR and SecureBio, to help assess its AI risks and safety practices, according to CNBC. The report's details focus on METR, which has been placed in the highest of three new evaluator tiers that Anthropic is developing.
Under the tier system, first-tier evaluators can conduct narrowly scoped risk investigations, second-tier evaluators can publish reports without Anthropic's approval, and third-tier evaluators receive employee-like access. METR is in the third tier: six of its employees have been onboarded, with access to Anthropic hardware, office spaces, and tools comparable to what the company's own risk assessment teams have.
The METR employees will analyze Anthropic's risk levels, incidents, and practices, then publish their findings. Anthropic said it is not funding METR's work, which may help preserve the evaluators' independence. The report does not provide details on SecureBio's specific role or access level.