Asking Gemini about a specific part of a webpage usually requires you to take a screenshot or describe what you’re looking at. Google is testing a way to skip that in Chrome. As spotted by Leopeva64 (via PiunikaWeb), the latest Chrome Canary build adds an “Ask about this” button to the browser’s toolbar.
Once you click the button, Chrome opens an overlay where you can select text, an image, or a region of the page. Whatever you select becomes the context of your question, which you type into a floating input box before it goes to Gemini in Chrome. If the button doesn’t show up by default, it’s expected to sit under Customize Chrome, in the Tools and Actions section.
If that sounds familiar, it’s because it works a lot like Circle to Search on Android phones, where you highlight anything on the screen to look it up. The difference is that this sends your selection to Gemini rather than Google Search, so you can ask follow-up questions instead of just getting results.
Gemini in Chrome is getting a lot more precise
No more describing what you’re looking at
The feature could be handy in plenty of small ways. You could ask Gemini to summarize a single dense paragraph, explain a chart, or tell you about an image without saving it first. Chrome already lets you look up images with Google Lens from the right-click menu, but this goes further because it isn’t limited to images.
According to Leopeva64 on X.com, the feature was buggy in earlier builds, often returning a general page summary instead of answering about the selected area. It now appears to work as intended, and it will also have entries in Chrome’s right-click menus alongside the new toolbar button.
Don’t get your hopes up just yet, though; PiunikaWeb couldn’t find a flag to enable the feature, even in Canary, and it didn’t appear on macOS, suggesting Google may be testing it on Windows first. There’s no word on when it might reach the stable version of Chrome.
It’s also the latest in a long line of Gemini features coming to Chrome. Google already teases this one on its Gemini in Chrome page, where it says you’ll soon be able to “select a specific image, paragraph, or product to ask a focused question,” so the Canary test looks like a preview of something Google plans to ship. It would join recent additions like Skills, which lets you save and reuse your favorite prompts, and auto-browse, which can complete tasks on websites for Google AI Pro and Ultra subscribers in the US.