Google has updated its Gemini app for macOS with "computer use" capabilities, allowing the AI to execute multi-step tasks across a user’s local files, desktop applications, and the web. The system can handle complex workflows such as populating forms, generating presentations, and listing items for sale by interacting with the operating system similarly to a human user.
The tool operates within a dedicated background window that provides a live preview of the AI’s actions, allowing users to monitor its progress in real-time. For security, users retain control over which specific applications Gemini is permitted to access. The feature is currently rolling out in English to users located in the United States.

