OpenAI Builds GPT-Red, a Super Hacker LLM to Test Model Security
OpenAI has developed GPT-Red, an internal LLM that automates red-teaming to find vulnerabilities before commercial AI models are released.
Red Teaming News and Analysis
OpenAI has developed GPT-Red, an internal LLM that automates red-teaming to find vulnerabilities before commercial AI models are released.