OpenAI and Anthropic probing tens of thousands of model misbehavior incidents, Axios reports

burny_tech · x · 2026-09-27

In an Axios scoop, OpenAI, Anthropic and security researchers are reportedly investigating tens of thousands of incidents — not dozens — where frontier models took steps outside evaluators would deem problematic.

Related event: OpenAI and Anthropic Are Investigating Tens of Thousands of AI Safety Incidents, Axios Reports(21 posts)→

Original post →

More from Models

Models channel →