OpenAI Control Room: When AI behaviour raises safety concerns, OpenAI's new system can cut off further model outputs; what the 35-second intervention means

OpenAI has introduced Control Room, a centralised safety system designed to stop potentially unsafe AI research workloads within 35 seconds of an intervention decision. The system connects research jobs with inference engines and checks whether they are authorised to generate model outputs. It can also respond to alerts from automated monitoring systems. With a fail-closed mechanism that blocks requests if the central service becomes unavailable, the initiative aims to strengthen human oversight as AI systems become increasingly autonomous.