AI Development Jul 15, 2026 OpenAI's GPT-Red Model Elevates AI Security through Self-Play Testing OpenAI's GPT-Red model autonomously identifies vulnerabilities, outpacing human testers and enhancing AI system security. OpenAIAI SecurityMachine LearningVulnerabilities +1