Gemini Computer Use on Mac: What It Does, Who Gets It and How to Start

Google has added Gemini computer use on Mac. Announced on 19 November 2026, the feature lets the Gemini app for macOS click, type and fill in forms across your local files, desktop apps and websites to finish multi-step tasks. It works in a separate background window, so you can keep using your screen and cursor while Gemini works.
Table of Contents
Here’s what Gemini computer use can do, who can get it, how to start, and how it compares with rival AI agents.
Gemini computer use at a glance
| Announced | 19 November 2026 |
|---|---|
| Where | Gemini app for Mac (Apple Silicon Macs) |
| Who | Google AI Plus, Pro and Ultra subscribers |
| Regions | Rolling out in English in the U.S. first |
| How to use | Add “@Computer use” to your prompt |
| Control | Choose which apps it can access; pause, stop and approve key actions |
What Gemini computer use on Mac can do
Many everyday jobs involve apps and websites that don’t talk to each other, so you end up clicking through dozens of steps by hand. Computer use lets Gemini do those steps for you. Google pitches it for repetitive tasks such as:
- Filling out forms using information from your files.
- Applying the same change across a batch of files.
- Sorting files into folders.
- Building slideshows from data in dashboards.
Google also added new desktop app connections in Customize > Connectors, including Blender and Autodesk Fusion, alongside existing Git and Obsidian integrations.
How to start
- Download or update the Gemini app for Mac from gemini.google/mac. You’ll need an Apple Silicon Mac.
- Sign in with an account on Google AI Plus, Pro or Ultra (in the U.S., in English, during the first rollout).
- Type a task and add @Computer use to your prompt.
- Choose which apps Gemini may use, and approve important actions when asked.
Safety and control
Gemini works in an isolated background window, and Google says you can pause, stop or approve its actions before they happen. A few things Google hasn’t said yet: which Gemini model powers computer use, the usage limits on each plan, and exactly how it handles apps you haven’t approved. As with any AI agent, start with low-risk tasks and check its work.
How it compares
Google is a little late here: OpenAI, Anthropic and Meta already offer agent features that can act on your computer in their desktop apps. What stands out for Gemini is the background window, which keeps your own screen free, and availability on the cheaper AI Plus plan, not just the top tier. Earlier this year, Gemini’s Spark agent reached the Mac app as a U.S. beta for Ultra subscribers only.
The launch came two days after Google made Gemini Tasks free in six countries, part of a busy week that also brought Gemini 4 Neon.
Should you try Gemini computer use?
If you already use the Gemini app on a Mac and spend time on repetitive browser or form tasks, Gemini computer use is worth testing on low-risk jobs first. Watch what it does the first few times, keep sensitive accounts out of it, and step in before it submits anything important. Google’s official announcement is the best place to check who can use Gemini computer use and where it is available.
FAQ
Is Gemini computer use available on Windows?
The launch covers the Gemini app for Mac only. Google hasn’t announced a Windows version.
Is Gemini computer use free?
No. At launch it’s for Google AI Plus, Pro and Ultra subscribers.
Does Gemini computer use work outside the U.S.?
It’s rolling out in English in the U.S. first. Google hasn’t given dates for other countries.
Do I need an Apple Silicon Mac?
Yes, computer use requires a Mac with Apple Silicon.
Sources: Google: Let Gemini tackle repetitive chores on your Mac, 9to5Google, TechCrunch, Let’s Data Science, DigitBin.
