Google gives Gemini the ability to use your Mac for you

Glass Epoch DeskOriginal digestSource: TechCrunch AI

AI-assisted digest, checked against the original source before publishing. How we work

Gemini on macOS can now navigate native apps and the web to run multistep tasks for you, with a live preview and approval gates.
Google gives Gemini the ability to use your Mac for you
Photo: Myriam Jessier · free stock license

Google said Thursday it is adding computer use to the Gemini desktop app on macOS, letting the assistant operate directly on a user's Mac to carry out tasks on their behalf. The agent can navigate native Mac applications and browse websites, working across local files, apps, and the web to complete multi-step requests.

Google describes the workflow this way: after a user states what they want done, the agent runs in a live preview inside the chat while the user's own screen and cursor stay free for other work. As with comparable computer-use agents, users can pause, stop, or approve Gemini's actions before they execute — a permission model designed to keep a human in the loop on anything consequential.

Google's suggested use cases are largely repetitive or clerical: filling out forms, running the same operation across a batch of files, sorting files into folders by stated criteria, and assembling slideshows from local files or data in online dashboards. The pitch positions the feature as a way to offload tedious, multi-step work rather than single-answer queries.

Availability is narrow at launch. The feature ships initially in English for users in the United States, and it requires the Gemini app for Mac plus a Google AI Plus, Pro, or Ultra subscription. Google did not say when it would reach other markets or other platforms, leaving Windows support and international rollout as open questions.

Google is arriving after its rivals. OpenAI, Anthropic, and Meta have already offered agentic features in their desktop applications for some time, and Google itself only this week broadened the rollout of Gemini's "Tasks" feature, which lets the AI sign into websites, make purchases, or book appointments. The company has also just begun admitting early testers to its revamped "CC" agent, first announced in September, aimed at keeping families organized across email, calendars, chats, and tasks.

What to watch next is how quickly Google widens the footprint: whether computer use follows Tasks into more markets and onto Windows, how approval prompts hold up as the agent takes on riskier actions like purchases, and whether a subscription gate keeps the feature behind Google's higher tiers as rivals price agentic desktop capabilities differently.

Key facts

  • • Google brought computer use to the Gemini macOS desktop app on Thursday.
  • • Agent previews its work in chat while the user's cursor stays free.
  • • Launch limited to English, U.S. users, behind AI Plus, Pro, or Ultra.
  • • Google also expanded Gemini's "Tasks" feature rollout earlier this week.
  • • OpenAI, Anthropic, and Meta already ship desktop agentic features.

Why it matters

Builders now face a crowded desktop-agent field where Google is playing catch-up, so differentiation will hinge on approval UX and platform reach rather than raw capability. Expect the near-term battleground to be where these agents run — Windows, international markets — and how much of the workflow humans must sign off on.

Spotted an error? Send a correction.

Get the day's AI news in one email

The Daily Epoch · 60-second read · unsubscribe anytime

Related

Agents
TechCrunch AI2 min read

Superhuman pushes its Go assistant everywhere

Superhuman, formerly Grammarly, is making its Go assistant the default AI layer across Docs and Mail, adding cross-app context, 20 third-party agents, and proactive drafting.