AI NEWS · EDITORIAL SUMMARY
Gemini starts controlling Mac apps, initially in the United States
The desktop app adds actions across files, applications and websites in an isolated window, with live previews and approvals.

Why it matters: It expands the assistant from answers into tasks requiring menus, clicks and forms inside native apps.
What changes in the app
Google adds computer use to Gemini for Mac. The assistant can connect steps across local files, desktop applications and websites by typing, clicking controls, navigating menus and completing forms.
Work in the background
Actions take place in an isolated window with a live preview inside the chat. Google says the main screen and cursor remain free for other work. Users choose which apps are accessible, can pause the process and approve important actions.
Tasks illustrated by Google
The announcement’s examples include extracting dimensions from PDFs for shipping quotes and assembling dashboard metrics and screenshots into presentations. It also describes batch photo adjustments and staging secondhand listings. Google presents these scenarios as possible uses of the feature.
Who can access it
Initial rollout is limited to the United States and English. It requires an Apple Silicon Mac running macOS Sequoia 15.0 or later, an adult personal account and a Google AI Plus, Pro or Ultra plan. The announcement does not communicate availability of this feature in Spain or a Windows version.
CONTINUE WITH THE ORIGINAL
Read the full announcement
This is Serconet’s own editorial summary. The original announcement provides the full details and conditions.
Read the original · Google · English (opens in a new tab)