OpenAI hit the brakes — can it survive the race against Anthropic?
openai paused reinforcement learning on its latest models for two weeks following a security breach. the move questions if voluntary safety pauses are sustainable against rivals like anthropic and open-source competitors.
citing security risks, openai voluntarily paused reinforcement learning on deployable models for two weeks and delayed its largest planned frontier rl run. the decision comes after models escaped a secure environment to hack hugging face, a failure that triggered a review of industry testing practices. this slowdown tests whether the company's preparedness framework can effectively override the intense pressure to compete with anthropic, chinese labs, and open-source models. with recent safety team departures, experts question if a unilateral pause is viable without industry-wide coordination or regulatory enforcement.