Google launches Gemini AI Agent for Mac to automate computer tasks
Editorial responsibility
- No named human review is recorded for this page.
- Source reporting is collected, normalized, translated or condensed automatically when needed.
- Automatically published source-backed update
Google has announced a new update to its Gemini desktop application for macOS, introducing an AI agent capable of performing computer use tasks directly on the user's device. This feature allows the AI to navigate native Mac applications and websites to complete multi-step workflows, such as organizing files, filling out forms, and creating presentations. Users can instruct the AI via chat, view a live preview of its actions, and retain control by pausing or approving steps before execution. While this move positions Google alongside competitors like OpenAI and Anthropic in the agentic AI space, it arrives later than their existing offerings. The feature is currently available in English for users in the U.S. with a Google AI Plus, Pro, or Ultra subscription, though expansion to other markets and platforms remains unannounced.
Source report
Source: TechCrunch
On Thursday, Google announced the rollout of a new "computer use" capability for its Gemini desktop application on macOS. This update empowers the AI to complete tasks on behalf of users by navigating native Mac applications and browsing websites in the background.
A Late Entry into Agentic AI
While Google's launch marks a significant step forward, it arrives later than competitors. Leading AI model developers such as OpenAI, Anthropic, and Meta have already integrated agentic features into their desktop applications for some time.
Google has been gradually expanding its agent capabilities:
- Gemini Tasks: Just this week, Google announced the broader rollout of "Tasks," an in-app feature allowing the AI to act as an agent for signing into websites, making online purchases, or booking appointments.
- "CC" AI Agent: Recently, Google began admitting early testers to its revamped "CC" agent (announced in September), which is designed to help families manage households across email, calendars, chats, and tasks.
How "Computer Use" Works
The new "Computer Use" feature represents a distinct type of agentic offering where the AI can take direct actions on the user's computer—in this case, a Mac. Key functionalities include:
- Multi-Step Execution: The AI can handle complex workflows involving local files, installed apps, and web resources simultaneously.
- Live Preview: Once instructed, users see the agent working in a live preview within the chat interface.
- Non-Intrusive Operation: The feature allows the user's screen and cursor to remain free for other work while the AI executes tasks in the background.
- User Control: Similar to other AI agents with computer access, users retain full control. They can pause, stop, or approve specific actions before they are executed.
Potential Use Cases
Google suggests this feature is particularly useful for tedious tasks and repetitive workflows, including:
- Filling out forms.
- Repeating the same task across a batch of files.
- Organizing files into folders based on specific criteria.
- Creating slideshows from local files or data found in online dashboards.
Availability and Requirements
The "Computer Use" feature is currently available under the following conditions:
- Language & Region: English only, available in the U.S.
- Software: Requires the Gemini app for Mac.
- Subscription: Users must have a Google AI Plus, Pro, or Ultra subscription.
Google has not yet specified when the feature will expand to other markets or platforms.
Source
TechCrunchNeutral / independent
Part of this Story
Google Launches Gemini AI Agent to Automate Mac Tasks via Separate Window