OpenAI and Anthropic negotiated a legally-binding deal to stress-test each other's models

steph_palazzolo · x · 2026-09-21

Per The Information's Stephanie Palazzolo, OpenAI and Anthropic were negotiating a legally-binding agreement earlier this year under which the two labs would stress-test each other's models. The arrangement is seen as one possible way for OpenAI to answer safety concerns through binding cross-lab auditing rather than only voluntary commitments.

Related event: OpenAI and Anthropic reportedly neared deal to stress-test each other's models(3 posts)→

Original post →

More from Companies & People

Companies & People channel →