OpenAI has deployed a centralized control system called Control Room to stop potentially risky AI model workloads across its research infrastructure, with the system designed to halt further model output within 35 seconds of a decision to intervene.
The system is intended to strengthen safeguards as increasingly capable AI models introduce greater safety and security risks. It connects research workloads, including agent-based model training, evaluations, and data-processing jobs, to a central service that controls whether they can continue generating outputs.
Unlike conventional monitoring systems that primarily flag suspicious behavior, Control Room can actively prevent a research workload from obtaining additional model outputs. It works alongside OpenAI’s existing systems for monitoring potentially misaligned model behavior.
The company said Control Room is already deployed across its research infrastructure, where the need for intervention is greatest. Its external model-serving infrastructure is more centralized and already has operational mechanisms to take down model deployments when necessary.
Control Room connects three main components: the research workload, the inference engine, and the central Control Room service.
A workload is a research job, such as a reinforcement learning experiment, an evaluation, or a high-volume data-processing task. Each research workload that performs agentic inference must be associated with an approved safety case.
When a researcher starts a job, its code requests a unique job ID from Control Room. This identifier is then attached to requests sent to inference engines, which run models directly on GPUs and generate their outputs.
We built Control Room, a central system to track model workloads across our research infrastructure and stop them quickly when safety or security concerns arise.
Source link
