Google has started rolling out a new “Select from screen” feature inside Gemini in Chrome. It lets users highlight specific text or images from the current tab and send them directly into a Gemini prompt, making the assistant’s replies more grounded in what’s
For the first time, Gemini inside Chrome isn’t just reacting to what you type. It’s starting to look at what you’re looking at.
Google has begun rolling out a new “Select from screen” feature inside Gemini in Chrome. The idea is simple: users can highlight specific text or images from a webpage and send them straight to Gemini. Once you activate the tool. the selected content is attached to your Gemini prompt automatically—no need to manually describe what’s on the page.
The feature shows up inside Gemini in Chrome’s “+” menu. After it’s turned on. it works much like a built-in screenshot tool: you select any text or image visible in the current browser tab. and Gemini receives it as part of the conversation. Instead of trying to translate what you see into words. the assistant can be aimed at the exact part of the page you want help with.
Google is rolling the capability out as part of Chrome 149. though some users may need to restart their browser before the option appears. The change lands as part of Google’s broader effort to make Gemini feel less like a standalone chatbot and more like an assistant that understands context as you move through Chrome.
The rollout also comes on the same day as another major Gemini update from Google. Developers can now access computer use capabilities directly through Gemini 3.5 Flash. In practice. that means AI agents can see. reason. and take actions across browsers. mobile apps. and desktop environments without relying on a separate model. Google says the integration is meant to improve long-horizon tasks such as software testing. enterprise workflows. and other multi-step automation jobs.
Taken together, the message is hard to miss. One update reduces the friction of getting Gemini to understand your immediate screen context. The other expands what Gemini’s tools can do across devices and environments. The common thread is getting the assistant closer to the real work happening in front of you—less time explaining. more time completing.
Gemini Chrome select from screen Google AI Chrome 149 Gemini 3.5 Flash computer use AI assistants browser automation