To detect potential issues in its AI models and quickly stop them in their tracks when needed, OpenAI has built Control Room.
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.
It connects every research inference workload, including agentic sampling for model training and evaluation, data…
— OpenAI (@OpenAI) November 9, 2026
Control Room
This is a central system to track model workloads across OpenAI’s research infrastructure and stop them quickly when safety or security concerns arise. It connects every research inference workload, including agentic sampling for model training and evaluation, data processing and more.
The Control Room works together with OpenAI’s misalignment monitoring systems and halts model output within 35 seconds of a decision to intervene to stop or pause a job. The 35-second lag comes from a caching layer for inference authorisation, to support the required volume of traffic. The company is also working to drive this number down to zero so that it can end jobs instantly.
It has been built to keep working even when other systems fail, and to stop model output if Control Room itself goes down. Fail-closed is rolled out to all frontier model research workloads, and will finish rolling out across all remaining research workloads over the next few days.
How does Control Room work?
It works by integrating a few systems: the workload, the inference engine, and the centralised Control Room service.
- The workload is the job being run, such as a reinforcement learning experiment, an evaluation, a high-volume sampling or data processing job.
- The inference engine, or just engine, is the model itself. You can think of the engine as the brain of the model, running directly on GPUs. A workload normally uses one or more models, which might be running on one or more different inference engines. The same inference engine might be used by more than one different workload.
- The Control Room service tracks jobs and controls whether they are allowed to continue generating model output. Engines check with the Control Room service before fulfilling a request. If a job has been stopped, the service denies the request, preventing further output.
Control Room is integrated in the following way:
- When starting a new job, the workload code requests a new job_id from Control Room.
- Whenever the workload needs to sample tokens from a model, it makes a request to an inference engine running that model, passing the job_id. The engine requests permission from Control Room to fulfil that request, passing the job_id along to Control Room.
- Control Room checks whether that job has been terminated, and if so, it denies the inference request. The engine responds to the workload’s inference request with a refusal, stopping further model output from being produced for that job.
Further, Control Room and OpenAI’s monitoring system are integrated in the following way:
- If the monitoring system detects a potential issue, depending on the details and severity, the monitoring system may decide to stop the job.
- In this case, the monitoring system sends a message to the Control Room to stop the job, passing the job_id. Control Room records that the job is terminated.
- Next time the job with that job_id makes an inference request to sample more tokens from any model running on any engine, the request will be denied, stopping further model output.
