Gemini in Chrome Adds 'Select from Screen'; 3.5 Flash Now Has Computer Control Ability

Google is rolling out a new "Select from screen" tool for Gemini in Chrome. Users simply click the button next to the Gemini prompt box to highlight the current tab and freely select any text or image on the screen — the selection is automatically added to the conversation. The feature is gradually rolling out with Chrome 149; if you don't see it immediately, restarting the browser will enable it. This move greatly simplifies feeding screen information into the AI assistant, making it ideal for quick questions or analyzing on-page elements.

At the same time, Google announced that the Gemini 3.5 Flash model now comes with a built-in "computer use" tool. Developers can call this capability through the Gemini API, allowing the model to directly interact with interfaces — for example, automatically analyzing the Gemini app and returning a categorized list of features. This means the AI can not only understand screen content but also simulate user actions to perform complex tasks. The computer use feature for 3.5 Flash is now available to API users.

These two updates target regular users and developers respectively, but they point in the same direction: making AI integrate more naturally into users' digital environments. From screen capturing to active manipulation, Gemini is evolving from a Q&A assistant into a true digital agent. With the broader deployment of Chrome 149 and the opening of API capabilities, more automation scenarios lie ahead.