Priorities and principles for effective third party assessments
AI Summary
OpenAI outlines priorities and principles for rigorous, secure, and independent third-party AI safety assessments of frontier models and safeguards.
Discussion
Sign in to join the discussion.
Loading comments…