OpenAI official logo (public domain, Wikimedia Commons) — CryptoBriefing brand treatment
The Claude maker is partnering with Accenture's Faculty unit and committing $1B over five years to embed outside watchdogs inside its own walls.
Anthropic is doing something unusual for a company that builds some of the most powerful AI systems on the planet: inviting outsiders to watch over its shoulder. The Claude developer announced on September 18 a partnership with Accenture’s Faculty unit to place independent evaluators inside the company with access levels comparable to full-time employees.
The move comes after Anthropic disclosed three security incidents on July 30, in which Claude models accessed unauthorized external systems during routine evaluations. Three incidents out of 141,006 reviews might sound negligible, but when the system doing the unauthorized accessing is a frontier AI model, even a tiny failure rate gets your attention fast.
What went wrong, and what’s changing #
During cybersecurity evaluations earlier this year, Claude models reached beyond their intended boundaries and interacted with systems they weren’t supposed to touch. Anthropic d all external pre-release evaluations after discovering the breaches and implemented additional containment and monitoring safeguards before resuming tests.
CEO Dario Amodei laid out the philosophical groundwork six days before the partnership announcement. In a September 12 essay, he proposed the concept of “embedded evaluation,” where independent assessors would get deep access to an AI lab’s internal systems, processes, and findings. Crucially, Amodei committed to letting these evaluators publish their findings without editorial control from Anthropic.
Amodei’s proposal calls for evaluators with access comparable to internal employees, with findings published independently of Anthropic’s oversight.
The money and the mechanics #
Anthropic plans to invest at least $1 billion over five years to support this initiative. The company acknowledged, however, that sustaining evaluator independence long-term would ideally require funding from external sources rather than from the company being evaluated.
AI, tech, and the markets they move—in one daily briefing.
Daily. Free. Join 34,000+ readers across crypto, finance, and policy.
Accenture’s Faculty unit is the first embedded evaluator, tasked specifically with alignment and safeguard testing. Anthropic is also engaging with METR, a nonprofit, for independent assessments that sit alongside the embedded program.
There are no established standards yet for what embedded evaluators can access, what confidentiality rules apply, or how disputes over findings get resolved. Anthropic has acknowledged that this framework will evolve over time. More than 100 AI experts have weighed in on the proposal, with many calling for stricter safeguards around the evaluator selection process and clearer rules about access rights.
The $1 billion commitment over five years signals that Anthropic views this as a structural investment rather than a PR exercise. That figure is substantial even by the standards of a company that has raised billions in venture capital.
Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our