Google announced on 19 November that Gemini can operate native Mac apps and websites through its macOS desktop application, TechCrunch reported. The feature lets Gemini carry out tasks across files, applications and the web after a user gives it instructions.
The agent’s actions appear in a live preview inside the chat, leaving the Mac’s screen and cursor available for other work. Users can pause or stop the agent, and approve actions before Gemini takes them.
Organising files into folders by specified criteria, filling out forms and making slideshows from files or online dashboards are among the uses Google suggests for the feature.
The launch follows a broader rollout of Gemini Tasks, which can help with website sign-ins, purchases and appointments. Separately, AI Affairs reported that free Gemini task automation reached six countries. Mac computer use is initially available in English in the US and requires a Google AI Plus, Pro or Ultra subscription.