Google is adding computer use to its Gemini desktop app for macOS. The feature lets the AI navigate Mac apps, work with local files and browse websites to complete tasks for users.
Sarah Perez reports for TechCrunch that the feature is initially available in English in the United States. Users need the Gemini app for Mac and a Google AI Plus, Pro or Ultra subscription.
After a user gives Gemini an instruction, the agent can carry out tasks involving several steps across files, apps and websites. A live preview in the chat shows its progress. The user’s screen and cursor remain free for other work.
Users can pause or stop the agent and approve actions before they happen. Google suggests using the feature for repetitive work, such as filling out forms or making the same change across a batch of files. It could also organize files into folders based on chosen criteria or create slideshows using files and information from online dashboards.
Part of a broader agent push
Computer use is one of several ways Google is expanding Gemini’s ability to act on users’ behalf. The company also announced a broader rollout of Gemini Tasks, an in-app feature that can help users sign in to websites, make purchases or book appointments.
Google has recently started admitting early testers to CC, a separate AI agent designed to help families organize email, calendars, chats and tasks. The company has not said when Gemini’s computer use feature will reach other markets or platforms.
Stay up to date
AI for content creation: the latest tools, tips and trends. Every two weeks in your inbox: