September 18, 2026
AI models will be tested from the inside: Anthropic and Accenture plan to invest at least $1 billion each
On 18 September, Anthropic announced a partnership with Accenture for independent model assessments. The work will be led by Faculty, Accenture's specialized AI unit: teams will evaluate models, look for vulnerabilities and test safeguards. Each company plans to invest at least $1 billion in this area over five years.
Previously, outside evaluators examined finished models from the outside. Now Anthropic wants to let independent reviewers inside the lab with employee-level access: they will be able to monitor model training and release, review safety decisions and speak with employees.
The rules are still being written. This setup assesses not only the model itself, but also the company's decisions around its training and release. Anthropic will fund Accenture's work directly, while shared rules for access and reporting by such evaluators do not yet exist.
Anthropic promises to announce other evaluators in the coming weeks.
