Google expanded Gemini's capabilities on macOS with a new computer-use feature that allows the AI assistant to interact directly with desktop applications, local files and websites. Announced on November 19, 2026, the update allows users to perform multi-step tasks.
With this capability, users can describe a task in natural language and ask Gemini to complete it. The AI can click buttons, type information, navigate menus, scroll through applications and fill out online forms.
The new computer-use capability supports several practical workflows. Google said the feature operates through an isolated background window. This allows Gemini to perform tasks without taking control of the user's primary screen or cursor.
Users can continue working on other applications while the AI handles the requested activity. A live preview within the Gemini chat interface also lets users monitor its progress.
For instance, users can ask Gemini to batch-edit photographs using desktop design software or prepare complex 3D scenes. It can also extract information such as weights, dimensions and addresses from local PDF documents and use those details to complete shipping quotation forms.
Giving an AI assistant access to desktop applications also introduces security concerns. Google also included permission controls that allow users to decide which applications Gemini can access.
The computer-use feature is initially rolling out in English in the United States. Users need an Apple Silicon-powered Mac running macOS Sequoia 15.0 or later.
Access is available to eligible personal Google Account holders aged 18 and above with Google AI Plus, Pro or Ultra subscriptions. The feature is not currently available for work or school accounts.
The update marks another step in Google's efforts to make Gemini an AI agent capable of executing tasks rather than simply responding to prompts.
Also Read: Google Gemini 4 Neon Launched with Advanced AI and Coding Features