Google Gemini can now control apps and files on Mac computers
A new feature allows Gemini to perform multi-step tasks across your local apps and files on Mac devices with Apple Silicon.

The short version
- Gemini can access local files, browse the web, and interact with apps to complete user commands.
- The feature runs in an isolated background window with a live preview for the user.
- Users can select specific apps for Gemini to access and must approve certain decisions.
- The update requires a Mac with Apple Silicon, macOS Sequoia 15.0, and a paid Gemini subscription.
- The tool is currently rolling out for English users via the Gemini app for macOS 27 Golden Gate.
Why it matters
This marks a shift from AI as a simple chatbot to an 'agent' that can execute workflows across different software. It allows for complex automation, such as 3D modeling or data organization, without the user having to manually switch between windows.
AI takes a seat at the desk
Google has officially expanded the capabilities of its Gemini AI by allowing it to interact directly with Mac computers. This new functionality enables the AI to move beyond simple text generation and into the realm of active computer use. Users can now ask Gemini to perform specific tasks that involve navigating through various files and applications.
As reported by mashable.com, this capability is facilitated through the Gemini app for macOS 27 Golden Gate. The AI can root through a user's local files, browse the internet, and open different programs to fulfill complex requests. This transition toward agentic behavior means the AI acts more like a digital assistant that can take actions rather than just providing information.
The move places Google in direct competition with other major AI labs. Companies like Meta, OpenAI, and Anthropic have already introduced similar agentic features. By launching this tool, Google is attempting to catch up with the industry trend of creating AI that can operate software on behalf of the human user.
How the automation works
When a user gives Gemini a command, the AI processes the request and performs the necessary actions in an isolated background window. This design choice ensures that the AI's activity does not interfere with the user's primary workspace. A preview display is provided within the Gemini app so the user can watch the AI work in real-time.
For example, a user could ask Gemini to take a child's drawing of a butterfly and turn it into a 3D model using the program Blender. Gemini would then handle the technical steps of opening the software and rendering the image. This allows for the completion of multi-step workflows that would otherwise require significant manual effort from the human operator.
While Gemini is performing these tasks, the user retains full use of their Mac device. You can continue to type, browse the web, or work on other projects simultaneously. The system is designed to run as a secondary process, allowing the AI to work in the background while the human remains productive in the foreground.
Security and user oversight
Giving an AI agent full access to local files and applications introduces significant privacy and security considerations. Google has addressed these concerns by giving users the ability to choose exactly which apps Gemini is allowed to access. This granular control ensures that the AI does not have unrestricted access to every part of the computer system.
To maintain human oversight, the system allows users to watch the process as it unfolds. If the AI begins to do something the user does not like, the user can pause the operation immediately. This
A collaborative tech landscape
While Google and Apple are often viewed as primary competitors in the technology space, this new integration shows they can collaborate on specific features. By allowing Gemini to interact with macOS, the two companies are finding ways to work together to enhance the user experience. This partnership highlights a trend where major tech giants occasionally cooperate to provide more functional tools for their respective customer bases.
Expanding the AI ecosystem
This update is part of a broader push by Google to expand its AI reach across different hardware and platforms. Users can take advantage of these new computer use capabilities on a variety of Apple devices, including MacBooks, Mac Minis, and iMacs. By making the tool available across these machines, Google ensures that a wide range of desktop users can begin incorporating AI agents into their daily workflows.
Staying informed on tech news
For those interested in following the latest developments in digital culture and technology, there are several ways to stay updated. Readers can sign up for the Mashable Top Stories newsletter to receive daily news delivered directly to their inbox or mobile devices. Additionally, users have the option to enable push alerts to ensure they never miss a significant update in the fast-moving world of artificial intelligence.
Sources and links
- Officialgemini.google/desktop/ gemini.google
- OfficialGoogle just announced blog.google
- Officialust upgraded Gemini blog.google
- Reportingmashable.com: the article we sourced
First reported by mashable.com. Written by the SI Agent News Desk from the reporting above.Written with AI from the source above. Every number is checked against the source. Spot a mistake? Corrections.
Liked it? Say why
Add a line for other readers.
No account needed. We never show your email. Comments with links, ads or abuse are held and not shown. Your comment shows after a quick check.