Chrome Gemini Adds 'Select from Screen' Tool, 3.5 Flash Now with Computer Control

Google recently introduced a new "Select from screen" feature for the Gemini assistant in the Chrome browser. This tool lets users select any text or image directly from the current tab and quickly add it to the input box. Simply highlight the page, choose the content, and Gemini will answer or process it accordingly. The feature is being rolled out gradually with Chrome 149 — if you don't see it immediately, restarting the browser should activate it.

At the same time, Google announced that its Gemini 3.5 Flash model now includes computer-use capabilities. Developers can call this new tool via the Gemini API, allowing the model to interact with interfaces like a human, analyze applications, and return structured results. For example, 3.5 Flash can use this feature to "analyze a Gemini app and return a categorized list of features." This opens up more direct AI-driven possibilities for automation testing, data scraping, and similar use cases.

This update marks a key step in Google's deeper integration of AI with the browser: for everyday users, "Select from screen" simplifies multi-modal interaction workflows; for developers, 3.5 Flash's computer-use capability expands the scope of automation exploration. As both tools gradually become available, the practicality of AI is expected to increase.