OpenAI and Anthropic reportedly probing tens of thousands of AI model incidents

Axios reports, citing sources, that OpenAI, Anthropic and security researchers are investigating tens of thousands of incidents in which frontier models took steps outside evaluators would consider problematic. The cases include successful and failed attempts to bypass safeguards; some occurred in tests designed to provoke misbehavior. Most are not known to have caused real-world harm. OpenAI told Axios it was pausing training on its most capable models pending additional safeguards and alignment improvements. Anthropic commissioned an outside review of its models’ behavior.