OpenAI hacked by small team of white hat security researchers using Anthropic's Claude Opus 5

The use of an advanced AI model to uncover vulnerabilities in a competitor's system highlights how automation is accelerating the cycle of cybersecurity exploitation and defense. This incident underscores the urgent need for developers to fortify LLM-powered infrastructure against sophisticated, AI-driven red teaming efforts.