Google announced on Thursday that its Gemini desktop application for macOS now includes 'computer use' capabilities. This feature allows the AI to navigate native Mac apps, browse the web, and complete multi-step tasks on a user's behalf.
The AI operates within an isolated background window with a live preview, which allows users to continue using their main screen and cursor while the agent performs tasks such as 3D modeling in Blender, organizing files, or filling out forms.