Beating AI News Flash: Gemini Mac client officially adds Computer Use functionality. Gemini can now directly operate software and websites on the computer, clicking buttons, typing text, switching menus, filling out forms, and completing complex cross-application tasks on its own.
Google demonstrated several practical uses: having Gemini operate design software to batch color-correct photos; extract dimensions, addresses, and other information from local PDFs to automatically fill out logistics quotation forms; or analyze clothing photos in a folder, look up secondhand prices, and prepare product listing materials.
Gemini operates in a separate background window, without occupying the user's main screen and mouse. Users can continue using their computer normally while watching the AI's operations through a live preview, pausing at any time, and specifying which applications it can access.
The new feature is currently rolling out gradually to paid Google AI Plus, Pro, and Ultra users in the United States only, and requires an Apple silicon Mac running macOS 15 or later.
