OpenAI Publishes Safety Case Guidelines for Frontier AI Training

This establishes formal safety protocols for frontier AI training, advancing governance and risk mitigation in the field.

Auto-scored: Publishing safety guidelines is a mitigating action with no direct risk, representing a real but partial step in AI governance.