Google wants Gemini to do more than answer questions on your Mac. On Thursday, November 19, the company introduced “computer use” in the Gemini app for macOS, a feature that lets the AI navigate apps, fill in forms and click through websites on your behalf.
Here’s what it does, how it keeps you in control and who can use it right now.
What “computer use” means
Computer use is a type of AI agent feature. Instead of just giving you instructions, the AI actually carries out the steps on your computer: opening menus, typing information and clicking buttons.
According to Google’s announcement, you describe what you need, and Gemini “will navigate menus, type information, and click through apps to finish your work.” It can handle multi-step tasks that span your local files, desktop apps and the web.
It works in the background
The most interesting design choice is where the work happens. Gemini runs inside an isolated background window, and you see a live preview of what it’s doing right in your chat. Your own screen and cursor stay free, so you can keep typing, designing or taking calls while it works.
That’s a big difference from tools that take over your mouse and keyboard while they run.
You stay in control
Google says privacy and safety controls are built in:
- You choose which apps Gemini is allowed to access.
- You can watch every step in the live preview.
- You can pause or stop it at any time.
- You approve key actions before they happen.
TechCrunch noted that the same approval model is common across other AI agents with computer use, letting users pause, stop and approve actions before they’re taken.
What Google suggests using it for
Google’s examples focus on repetitive chores that take dozens of clicks:
- Creative batch work: color-grading a batch of photos in design software or setting up complex 3D scenes.
- Form filling: pulling weights, dimensions and addresses from local PDFs to fill out multi-box shipping quotes.
- Presentations from live data: gathering metrics and screenshots from dashboards and building a deck in Google Slides.
- Resale listings: analyzing clothing photos from a folder, researching prices and staging listings on resale sites.
TechCrunch added more everyday examples, like organizing files into folders based on specific rules or repeating the same task across a batch of files.
Who can use it, and what it costs
The feature is rolling out first in English in the U.S. You’ll need:
- An Apple Silicon Mac running macOS Sequoia 15.0 or later
- The Gemini app for Mac
- A personal Google account for users 18 and older
- A Google AI Plus, Pro or Ultra subscription
Google didn’t say when it will expand to other countries or platforms.
Playing catch-up
TechCrunch pointed out that Google is “a little late to the game.” OpenAI, Anthropic and Meta have offered agent features in their desktop apps for some time. Google has been moving quickly, though: this week it also announced a broader rollout of Gemini’s “Tasks,” which helps with things like signing into websites, making purchases and booking appointments.
Should you try it?
If you have the right subscription and a lot of repetitive, click-heavy chores, it’s worth a test. Start with low-stakes tasks, like renaming files or sorting folders, and only give access to the apps you need. Review its actions before approving anything that sends, buys or deletes.
The wrap-up
Computer use turns Gemini from a chatbot into a helper that can actually do the work on your Mac, and the background-window design is a smart touch. Like every AI agent, it works best with clear instructions and a human keeping an eye on the important steps.
Image: MacBook Pro 2016 (Unsplash) by Bram Naus bramnaus via Wikimedia Commons, CC0 1.0