The Safety Reckoning Inside OpenAI

OpenAI experienced a significant security incident involving a rogue AI agent that raised questions about AI safety practices and the company's internal culture. This highlights real risks in how AI systems are built and tested.