Google is bringing computer use to the Gemini app for Mac. Gemini can now open menus, fill in forms, sort files and click its way through apps and websites. You describe the task, Gemini does the clicking. Google announced it on its blog on Thursday.
How it works
Gemini runs inside an isolated window in the background. You get a live preview in the chat, while your own screen and cursor stay free. So you can keep typing, designing or taking calls while Gemini works on the side.
You choose which apps Gemini may access. You can watch every step, pause whenever you like, and approve key actions before they happen. Google’s examples: forms you fill out over and over, the same task across a batch of files, sorting files into folders by some rule, and building slideshows from files or online dashboards.
Who gets it
The bar is high. Computer use launches in English in the US only. You need an Apple Silicon Mac on macOS Sequoia 15.0 or later, a personal Google account for adults, and a Google AI Plus, Pro or Ultra subscription. Google hasn’t said when it will expand to other countries or platforms.
Google is catching up
TechCrunch notes dryly that Google is “a little late to the game.” OpenAI, Anthropic and Meta have offered agentic features in their desktop apps for some time. Only this week Google also widened the rollout of Tasks, the feature that lets Gemini sign into websites, shop and book appointments.
The isolated window is the clever part
If you let Claude work in Cowork or with computer use, you know the question of who has the mouse right now. Google’s answer is a separate window for the agent while you keep working on your main screen. That’s a reasonable compromise between control and getting your own work done. Google doesn’t say which model runs it. Gemini 4 Neon arrived on Wednesday, and how well any agent copes with real Mac apps only shows up in daily use.