Google’s Gemini app for macOS now includes a computer use feature, allowing the assistant to carry out multi-step tasks across local files, desktop applications, and the web, the company says. The capability is aimed at Mac users who want help completing work that spans several apps rather than a single chat window. Google says the feature is rolling out now in English to users in the United States.
According to Google, computer use can assist with tasks such as accelerating workflows, completing forms, building presentations, and writing listings for items a user wants to sell. The company positions the tool as a way to hand off sequences of steps that would otherwise require switching between programs and browser tabs. The feature requires access to the user’s applications to perform that work.
Google says users control which apps Gemini may access, and that they can watch the assistant operate. Gemini runs in its own background window that displays a live preview within the chat, according to the company, so users can follow its progress as it works. That design gives users visibility into the actions the assistant takes on their machine.
The launch comes ahead of Apple’s planned restrictions for apps that request full disk access on Mac, which the source article links to potential risks posed by AI agents with computer use capabilities. A preview of those settings arrived with this week’s macOS 27.3 beta. Gemini’s rollout therefore lands just before the new Apple requirements take effect.
The macOS app is the latest expansion of Gemini’s computer use functionality, which lets the assistant act on a user’s behalf rather than only answering questions. Google’s description emphasizes local files and desktop software alongside the web, extending the assistant’s reach beyond browser-based tasks. The feature is available in English for US users at launch.
More software news from TechManNews.






