On this page · 4 sections ▾
Google announced on November 19 that computer use is rolling out in the Gemini app for Mac. It lets the assistant click, type, navigate menus and fill out forms across local desktop apps and websites to finish multi-step tasks. It is available in US English to Google AI Plus, Pro and Ultra subscribers.
Key facts
- Gemini works in an isolated background window with a live preview in the chat, so the user's screen and cursor stay free, according to 9to5Google.
- It requires an Apple Silicon Mac running macOS Sequoia 15.0 or newer, and users must be 18 or older with a personal Google account, per Android Authority.
- To trigger it, add @Computer use to a prompt.
- The rollout is in English in the US. TechCrunch reports Google did not say when other markets or platforms would follow.
- Google's examples include pulling weights and addresses from PDFs into shipping quote forms, building Slides decks from dashboard data, batch-editing photos and staging resale listings.
What changes for Mac users
Gemini's existing Task feature, previously called Gemini Spark, could only reach local files and connected apps such as Gmail or Google Calendar, according to 9to5Google. Computer use extends that to operating the apps themselves. New "Desktop" connectors, including Blender and Autodesk Fusion, appear under Customize > Connectors next to existing Git and Obsidian integrations.
Android Authority describes a Google demo in which a parent asks Gemini to turn a child's butterfly sketch into an animated 3D model in Blender.
Safety controls and risks
Users choose which apps Gemini can use, can stop it with one click and must approve sensitive actions. Google will pause for logins and payments, per 9to5Google.
The Next Web adds that Google warns the agent can make mistakes, including permanently deleting files or sharing them with others. It is designed to ask permission first and can only see folders the user connects. Backups of changed files are deleted after 24 hours or when a new task starts. The Next Web also reports the feature is not available in Europe or the UK.
The Verge notes the launch comes as Apple prepares new restrictions for apps requesting "full disk access" on Mac because of risks from AI agents. A preview of those settings arrived in the macOS 27.3 beta this week.
What to watch
TechCrunch frames Google as late to desktop agents, noting that OpenAI, Anthropic and Meta have offered agentic features in their desktop apps for some time. Google has moved quickly in recent days, with Gemini Tasks broadening its rollout shortly before this launch.
Two things are worth tracking: how Apple's full disk access rules affect agent apps once they leave beta, and whether Google extends computer use beyond US English, other platforms or free tiers. Google has given no timeline for any of these.




