⚡️ JUST IN: OpenAI says some previously released GPT-5 models were accidentally exposed to limited chain-of-thought grading during RL training, but found no clear evidence of major monitorability degradation.

post-image
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pin