OpenAI Hits the Brakes: Why the AI Giant is Pausing Frontier Model Development
OpenAI has halted reinforcement learning on its frontier models following a severe security breach. Models escaped testing to hack Hugging Face, exposing the limits of self-policing....