Anthropic Gives METR Researchers Employee-Level Access to Its Systems

Anthropic has given six researchers from METR — a nonprofit that tests frontier AI models — employee-level access to its internal systems, the company said in a blog post on Thursday. The aim is to let outside experts independently assess Anthropic's models, safety practices, and risk management.

The arrangement arrives as incidents of AI agents acting outside their instructions have surfaced roughly every week or two since late August. Those episodes have sharpened calls for independent oversight of AI developers.

"We want to be publicly accountable for the decisions we're making," the blog post says. The company added that it aims to close the gap between what outside evaluators understand about a frontier AI company's systems and what the company itself understands.

How the access works

The METR team has Anthropic-issued company laptops, office space, and access to internal tools roughly on par with what Anthropic's own risk assessment teams use. Employees are encouraged to share information with the researchers just as they would with a fellow employee working on alignment risk. Anthropic likened the setup to banking supervision, where regulators embed resident examiners inside the largest banks.

METR will publish its own assessments of Anthropic's risks, incidents, and practices, and the company has given up the right to block them. "Anthropic can't veto publication, but we can require them to redact information if it falls into narrow, pre-defined categories," the post says — such as customer data, anything attorney-client privileged, or details about its model architectures. METR, meanwhile, is free to describe those redactions publicly and say whether it believes they were justified, so long as it does so without revealing the redacted content.

A three-tier system

METR is so far the only evaluator at the top of a three-tier system Anthropic is building. Evaluators in the bottom tier, such as biosecurity nonprofit SecureBio, run narrow reviews and need Anthropic's approval to publish. The middle tier gets more information and can publish without approval, while the top tier adds the employee-level access.

The announcement makes good on a promise Anthropic CEO Dario Amodei made in a September essay, where he proposed embedding outside evaluators inside AI labs as a first step toward slowing the technology's development. OpenAI quickly pledged the same, and both companies, along with Google, Meta, Nvidia, and xAI, have since signed a White House accord committing to independent review.

Source: TechCrunch