Google has added a “computer use” feature to its Gemini app for macOS, allowing it to perform multi-step tasks across local files, desktop applications, and the web. The assistant clicks, types, navigates menus, and fills out forms in a separate background window, with a live preview shown in the conversation. Users can grant access app by app, monitor the actions, and pause or stop an operation at any time. Launched on November 19, 2026, the feature is available in English to US subscribers to Google AI Plus, Pro, or Ultra. It follows similar tools introduced by OpenAI and Anthropic, with Google emphasizing user control and visibility as key differentiators.
Source: Read the original article


