Safety · OpenAI Blog ·

GPT-Red: Unlocking Self-Improvement for Robustness

GPT-Red: Unlocking Self-Improvement for Robustness

OpenAI describes GPT-Red, an automated red-teaming system that uses self-play to test and improve AI safety, alignment, and resistance to prompt injection. The system is presented as a method for generating robustness evaluations.

Read the full story at OpenAI Blog →