Google is bringing computer use to its Gemini app for macOS, allowing the assistant to operate across desktop applications and websites in a separate background window without taking over the user’s screen, The New Stack reported.
The feature includes a live preview in the chat interface, letting users monitor activity, pause execution, and approve actions when necessary.
The update lets Gemini click and type for the user to finish multi-step tasks such as moving through menus and filling out forms, according to 9to5Google. Users can prompt the assistant by typing “@Computer use” to execute tasks while their main screen and cursor remain free.
Google lists several practical use cases for the Mac tool, including batch color-grading photos in design software, setting up 3D scenes, extracting dimensions from local PDFs to fill shipping quotes, and building presentation decks in Google Slides from live regional dashboard metrics. In the app’s Customize settings under Connectors, new desktop applications like Blender and Autodesk Fusion appear alongside existing Git and Obsidian integrations.
With computer use in the Gemini app for Mac, you can now hand off tasks across your folders, apps, and the web, while your screen and cursor stay yours. Watch Gemini click, scroll, and type in a live background window, turning a quick sketch into a rigged 3D model in Blender.
Google announced the Mac expansion on Thursday, arriving just three days after Apple introduced tighter permission controls for macOS automation. The macOS 27.3 beta 2 changes split Device Control and Data Access permissions into two categories: System & App Protection Override and Accessibility.
Apps requesting broad system access face explicit permission prompts, allowing users to approve individual requests rather than granting ongoing access. Developers building desktop agents must adapt to least-privilege access rules, as users can restrict or revoke permissions while an agent runs.

Google has not explained how Gemini will operate under these stricter macOS restrictions or detailed how its background environment isolates from the operating system. Furthermore, the company has not clarified whether the consumer Mac app shares the execution architecture of Google’s developer-facing computer-use API, which debuted with Gemini 2.5 in October 2025 and integrated into Gemini 3.5 Flash.
The computer use feature is currently rolling out in English for personal Google accounts belonging to users aged 18 and older in the United States, The New Stack reported. Running the tool requires an Apple Silicon Mac equipped with macOS Sequoia 15.0 or later.
Users also need a paid subscription tier to access the functionality. Eligible accounts must hold an active Google AI Plus, Pro, or Ultra subscription.
The assistant pauses for logins and payments, requiring manual confirmation for sensitive actions. Users can also stop execution at any time with a single click.
Desktop apps such as Blender and Autodesk Fusion join existing Git and Obsidian integrations within the Customize menu under Connectors.
Google has not stated whether it plans to open its background execution environment to third-party developers or if the consumer app uses the same architecture as its developer APIs.