OpenAI outlines technical controls and veto guidelines for frontier AI safety

No summary available