GPT-Red

Trends

• Frontier labs slow down and publicly report their own risks

Items

1. Jul 15

   OpenAI's GPT-Red automated red-teamer

   Trained by self-play RL, the red-teaming model finds prompt-injection attacks on 84% of held-out scenarios (versus 13% for human red-teamers) and was used to adversarially train GPT-5.6 Sol to roughly sixfold fewer failures on OpenAI's hardest direct-injection benchmark.