Skip to content

OpenAI dots can now use several of your computers at once

Comic-style developer stretches glowing neon cables between a laptop, a desktop PC, a build monitor and a server rack AI agents

OpenAI dots can now work across several of your computers. On November 9, 2026, OpenAI added support for connecting multiple local machines to the same dot. You choose which computer it uses for each task.

OpenAI's own example is a developer workflow. A dot can build and test an app on a dedicated machine while you keep coding on your laptop. Until now, a dot could use only one personal computer at a time.

What changed in OpenAI dots

OpenAI announced these always-on agents inside ChatGPT at DevDay on September 29. Each dot has its own cloud computer and browser and keeps working when your devices are off. Connecting a local computer gives it access to that machine's files, code and apps.

Now you can connect more than one. The setup is the same for every machine:

  1. Open your dot's profile in the ChatGPT app on that computer.
  2. Under Computers, find the machine and select Allow.
  3. Review the confirmation and select Allow access.

To add another computer, you repeat these steps in the ChatGPT app on that machine. The dot's cloud computer stays connected too. Access applies wherever you message the same dot, including from your phone.

There is no per-task dropdown. You tell the dot which machine to use. For independent tasks, OpenAI suggests asking it to start a separate thread on each computer. Each thread uses one computer at a time.

OpenAI's demo: one request makes a dot check an iOS build on a Mac Studio and fix backend bugs on a Dev Box at the same time. Video: OpenAI.

In the demo, the user's device list shows a laptop, a Mac Studio, a Dev Box, a MacBook Pro and a build server. One message sends an iOS build check to the Mac Studio and a Linear bug fix to the Dev Box. The dot starts both tasks and keeps the user posted.

The demo shows the idea, but the real use depends on what a dot can do on each machine.

What a dot can do on a connected computer

On a connected machine, a dot can create local Work or Codex tasks and continue existing local Codex tasks. It can use local skills, developer tools and desktop apps. If a website blocks the cloud browser, the dot can try the browser on your computer instead.

The dot coordinates this work from the cloud. A task can be managed there while it uses files and tools on your machine. In the desktop app, the Activity tab shows where each task runs, and you can step in with instructions.

A few rules apply to every connected computer:

  • the machine must be online with the ChatGPT app open;
  • an Offline status does not revoke access, and Revoke access removes it;
  • connecting a computer to Codex or turning on Work Sync does not give the dot access;
  • changing the dot's selected computer does not move a task that is already running.

For coding work that should not depend on any of your machines, a dot can also use a Codex cloud environment. Your computers can stay offline in that case.

Access to all of this still depends on your ChatGPT plan.

Who can use dots and what it costs

Dots launched on GPT-6 Astra. OpenAI has since added GPT-6.1 Terra to the models that power them, and a dot can use different models for different parts of its work.

At launch, dots were available on these plans:

  • Pro, for adults outside the European Economic Area, the UK and Switzerland;
  • Business Premium, at $100 per user per month billed annually or $125 billed monthly;
  • Enterprise, Edu and Healthcare, where admins must turn dots on. Billing for those workspaces starts on November 18.

Free, Go and Plus users do not have dots yet. The first dot is included at no extra cost on eligible plans. Chats with a dot do not count toward ChatGPT limits. Tasks it starts in Work or Codex count toward those products' limits as usual.

Local access works on Mac and Windows. OpenAI's notes for November 2–6 say dots now continue local tasks more reliably on Windows. In Enterprise workspaces, admins turn on local computer access for dots separately.

More machines also mean more places for an agent to act, and the controls have not changed much.

Limits of multi-computer dots

OpenAI's guide describes one Allow access step per computer. It mentions no per-machine permission levels, so every connected computer gets the same kind of access. The guide also names no maximum number of machines.

It does not explain what happens to a running thread when its computer goes offline mid-task. The demo comes from OpenAI, and no independent tests of two-machine setups have appeared yet.

The existing safeguards still apply. Dots follow your ChatGPT app permissions, automatic review checks some actions, and sensitive steps such as money transfers stay with you. OpenAI also warns that a completed run does not by itself confirm the result. Users should review the output and any reported errors.

Is multi-computer support in OpenAI dots worth using

For developers with a spare Mac or a build server, yes. A long iOS build or a test suite can run on another machine while your laptop stays free for coding. The dot keeps track of both jobs and reports back, even when you message it from your phone.

For everyone else, the update changes little. Most people have one computer, and the cloud computer already covers research and documents. The trade-off is access: each connected machine opens its files and tools to the dot with a single permission. Start with a dedicated machine that holds only the projects you want the dot to touch.

Sources

Portrait of Felix Laurent, author at thehype.news

Felix Laurent

News Writer at thehype. Felix is a journalist with experience in marketing. He covers artificial intelligence and emerging technologies, focusing on the ideas, innovations, and trends shaping the future of tech.

More articles by Felix Laurent →
On air · radio.thehype.news ↗ AI news radio — 24/7