Google announced on Thursday that computer use is rolling out in its Gemini app for Mac, allowing the AI to complete tasks on users' behalf, navigating native Mac apps and browsing websites in the background. Users describe a task, and Gemini works through it across local files, desktop apps and websites in a separate background window, with a live preview in the chat so the user's screen and cursor stay free.
The company is a little late to the game, as OpenAI, Anthropic and Meta have already offered agentic features in their desktop applications for some time. Google this week also announced the broader rollout of Gemini's "Tasks" feature, which helps users sign into websites, make online purchases or book appointments, and began admitting early testers to its revamped "CC" AI agent, announced in September, designed to help families stay organised across email, calendars, chats and tasks.S1
Users can choose which apps Gemini may use, pause it, stop it, and approve key actions before they happen. Google's examples include pulling weights and addresses from PDFs into a shipping quote form, turning dashboard data into a Google Slides deck, batch-editing photos, and preparing resale listings from pictures in a folder. Google warns that Gemini can make mistakes, including permanently deleting files or sharing them with others. It is designed to ask permission first and can only see folders users connect, and backups of changed files are deleted after 24 hours or when a new task starts.S1S2
The feature is available in English in the U.S. only and requires the Gemini app for Mac, a Mac with Apple Silicon, and a paying Google AI Plus, Pro or Ultra subscription. It is not available in Europe or the UK, and Google has not said when it will expand to other markets or platforms. Google first built computer use into Gemini 3.5 Flash for developers in June, and is also building Gemini into the operating system of its Googlebook laptops as rivals race to put agents on the desktop.S1S2