Google announced on Thursday that it is expanding the capabilities of its Gemini desktop application on macOS. The update introduces a computer use feature that allows the artificial intelligence to complete tasks on behalf of users. The system can navigate native Mac applications and browse websites in the background.
This launch places Google slightly behind competitors in the agentic AI space. Companies such as OpenAI, Anthropic, and Meta have already offered similar agentic features in their desktop applications for some time. The move follows Google’s broader rollout of Gemini Tasks earlier this week. That feature allows the AI to act as an agent for actions like signing into websites, making online purchases, or booking appointments.

Autonomous Navigation and User Control
The new computer use functionality enables the AI agent to take direct actions on the user’s computer. It handles multistep workflows that involve local files, applications, and the web simultaneously. After a user provides instructions, the agent works in a live preview within the chat interface. This design keeps the user’s screen and cursor free to continue working on other tasks.
Google emphasizes that users retain full control over the process. Similar to other AI agents offering computer use, users can pause, stop, or approve actions before they are executed. This safety measure ensures that the AI does not perform unintended operations on the system.

Google suggests the feature is particularly useful for tedious tasks and repetitive workflows. Potential use cases include filling out forms, processing batches of files, organizing documents into folders based on specific criteria, and creating slideshows from online dashboard information.
Availability and Subscription Requirements
The computer use feature is initially available in English for users in the United States. Access requires the Gemini app for Mac and an active Google AI Plus, Pro, or Ultra subscription. The company has not provided a timeline for expanding the feature to other markets or operating systems.

This release also coincides with the recent admission of early testers to try the revamped CC AI agent. Announced in September, the CC agent is designed to help families stay organized across email, calendars, chats, and tasks. The introduction of computer use marks another step in Google’s strategy to integrate more autonomous capabilities into its consumer AI products.
Source: TechCrunch

