GeethanTechGeethanTechTech, decoded daily.
LatestAISoftware & PlatformsCybersecurityStartups & VentureTech Business & Markets
All posts
AIabout 8 hours ago

Gemini for Mac adds computer use in a limited US rollout

By megan

Google says Gemini can work across Mac apps, local files, and websites in a separate background window. The English-language rollout is limited to eligible US personal accounts on paid plans.

Conceptual illustration of a human Mac workspace beside a separate abstract task window with document, app, and web shapes.

Google is rolling out computer use in the Gemini app for Mac. In its 19 November 2026 announcement, the company said Gemini can click, type, and navigate menus across local files, desktop apps, and websites to carry out multi-step tasks. This extends the Mac app from answering questions to taking actions in other software.

How it works

Google says the task runs in a separate background window with a live preview in the Gemini chat. The user’s main screen and cursor remain available. Examples Google gives include transferring information from local PDFs into web forms and assembling presentation material from dashboards. Those examples describe intended uses, not independently verified results.

The Mac computer-use help page says users must enable the feature and grant macOS Accessibility and Screen & System Audio Recording permissions. They can allow or block individual apps, view the task as it runs, and stop it. Choosing “Always allow” adds an app to a persistent allowed list; it does not mean Gemini asks before every subsequent use of that app.

Who gets access

Computer use is rolling out in English in the United States. Google limits it to people aged 18 or older using personal Google accounts with Google AI Plus, Pro, or Ultra. It requires an Apple Silicon Mac running macOS Sequoia 15.0 or later. Google’s help page says work and school accounts are not currently eligible for these Gemini tasks.

Where oversight matters

Google’s permission guidance describes one-time and saved permission choices for browser and app actions. Some sensitive actions can still require confirmation. Users should review the apps they allow, especially because Google warns that Gemini can make mistakes, access signed-in desktop sessions, and encounter untrusted content that attempts to influence actions across apps. Google also cautions that access to developer tools may permit broader file changes or shell actions. A live preview and permission controls help users supervise a task, but they do not establish that every result will be correct or reversible.

Continue with this topicAI
Browse categoryarrow_forward
Continue reading

Related posts

A blank desktop monitor behind a document, clock and stacked cards, with a green permission switch beside them.
AI2 days agoBy megan

Proton launches Lumo Desktop with connectors and scheduled tasks

Lumo now has desktop clients for macOS, Windows and Linux with connectors, schedules and Skills. Proton's setup guide says connector actions default to Auto-approve until changed.

Read article
A paper order slip and a generic payment token separated by a manual approval point.
AI6 days ago

ChatGPT Wallet begins limited US rollout with approval for every purchase

OpenAI’s new wallet lets eligible US Pro users approve purchases made by their dot. Merchants still control checkout, fraud decisions and order confirmation.

Read article
GeethanTech
GeethanTech
Tech, decoded daily.
Read
Latest postsAISoftware & PlatformsCybersecurityStartups & VentureTech Business & MarketsRSS feed
Connect
mailEmaillanguageWebsiteinLinkedInIGInstagramfFacebook@Xsmart_displayYouTube
From the publisher

Read all our publications: GeethanPost | GeethanTech

A chapter in the ElegantHumanity story

Powered byElegantArc