NewsOpenAI (official)Sep 22, 2026
Practical guidelines outline new approach to evaluating and governing frontier models
Practical guidelines for safety evaluation and governance of frontier models released. They address third-party assessment and transparency, boosting industry trust.
Why it mattersStandardization of evaluation criteria is important for regulation and audits; helps compliance.
Read the original →