Google Gives Gemini Control of Mac Apps, But Its New AI Agent Comes With Risks
Gemini's new Mac computer-use feature navigates apps, files, and websites in a separate window, but Google's own documentation warns about prompt injection.
Jose Antonio Lanz
Publisher Decrypt
Nov 20, 2026 at 6:16 PM UTC · Updated 6시간 전 · 4 분 소요

Key Signal
macOS 15.0+ Minimum Mac operating system
Last Updated
6시간 전
- Google launched computer use for its Gemini Mac app, allowing the AI to click through desktop applications and websites while users keep working.
- The agent runs in a separate virtual screen, with app permissions, live previews, and confirmation requests for sensitive actions.
- Google warns that malicious content, broad app access, and imperfect backups can still put user data and files at risk.
Google wants Gemini to do more than answer questions on your Mac.
The company launched a computer-use feature Thursday that lets its AI assistant open applications, navigate menus, fill out forms, and move information between local files and websites on a user's behalf.
“Starting today, computer use in the Gemini app for Mac helps you get multi-step tasks done across your local files, desktop apps, and the web,” Google announced. “Just describe what you need, and Gemini handles the clicking, typing, menus, and forms for you.”
The new capability runs in what Google calls an isolated background window, leaving the computer's main display and cursor available for the owner. Users can watch the agent's actions through a live preview inside the Gemini chat and interrupt a task if it goes off course.
Article Intelligence
Topics
Sponsored
AdNewsLayer Premium
Unlock deeper intelligence.
Ad-free reading, exclusive research, and real-time onchain insights.
Go Premium
