What happened
Google has added a feature called Computer Use to its Gemini app for Mac, The Decoder reported. It lets the AI carry out tasks that span several steps across desktop software, files stored on the computer and websites. According to Google, a user states what they want done, and Gemini then handles the clicking, typing and movement through menus and forms needed to get there.
The work takes place in a separate, isolated window running in the background, with a live preview shown in the chat, so the user's own screen and cursor stay free. Users decide which applications Gemini may access and can follow each step as it happens. They can also pause the process at any point and must approve important actions before they are taken.
Google gave several examples of how the feature can be used. These include editing batches of photos or 3D scenes, completing shipping forms with information pulled from local PDFs, and assembling presentations. Computer Use is rolling out in English for users in the US starting today, and the Gemini Mac app can be downloaded now.
Context
The launch moves Gemini beyond conversation and into directly operating software on a user's computer. It also places Google in competition with OpenAI, which has recently put its own focus on computer use capabilities. The report describes OpenAI as being well ahead in this area, which suggests that Google is working to catch up rather than leading the category.
The briefing
The week in tech, in one email.
The stories worth your time, every week. Unsubscribe anytime.
