Chrome Can Now Use Artificial Intelligence to Browse the Web.
Have you ever wanted to browse the internet but found the thought of typing a URL into the address bar too tedious? Now Google is ready to help. Today, the company announced a significant expansion of the existing Gemini feature in Chrome . Key features include a new look for this AI assistant, more integrated image editing tools, and, perhaps most impressive (but also terrifying), the launch of Auto Browse, which allows Gemini to take control of your browsing experience.
New sidebar look
Previously, Gemini in Chrome appeared in a small window at the top of the browser window, making it somewhat awkward to use, especially when switching between tabs. In Google’s update, it has been moved to a scrollable sidebar, which is slightly larger and won’t obscure other content. Instead, it will be positioned to the right of the webpage you’re viewing, making it easier to compare the answers Gemini provides with what you’re currently viewing or continue a conversation while switching between tabs. It will retain all the previous functionality, including the ability to reference multiple open tabs in queries. This is a small change, but it should improve usability.
Right click on any image to edit it in Nano Banana.
Google’s Nano Banana AI image generator is very popular right now, and new updates to Gemini in Chrome make it easier to use. Now, instead of downloading an image and re-uploading it to Gemini, you can edit it with Nano Banana with a simple right-click. Or, you can use natural language to start editing by opening the desired image on the screen and asking Gemini to edit it in the sidebar. Google claims this should work with virtually any image that can be opened in the browser.
During a demo, Google showed journalists this feature using the Google Photos library, but there’s no requirement to use only your own images. This immediately raised alarm bells, given that Elon Musk’s company, X, is currently at the center of a controversy after allowing anyone to use Grok to edit other people’s images directly on the social network without the owner’s permission. After some users began using the tool to create explicit content from other people’s photos, it was slightly restricted, but Google doesn’t seem concerned. When I asked about the security measures for this feature, a Google spokesperson told me the following: “We have clear policies prohibiting the use of our AI tools to create sexually explicit content, and our tools are constantly improving to comply with these policies. We’ve invested in security from the start and added technical safeguards to limit problematic results, such as violent, offensive, or sexually explicit content.”
The company hasn’t said anything about how users can use Nano Banana in Chrome to bypass copyright restrictions, but technically, the new update doesn’t actually add new features to Google’s AI-powered image generator; it simply makes it easier to access. Of course, the same applies to the recent Grok update, and making access easier can mean opening the floodgates, even with the best intentions.
Use AI to automatically browse pages in Chrome.
And finally, and most importantly: “Agent-based” is the latest buzzword in the AI world, and Google doesn’t want Chrome to fall behind. So now, instead of simply answering questions, Gemini can take control of your browser.
This feature is currently only available to Google AI Pro and Ultra subscribers , but starting today, those subscribers can ask Chrome to use the “Automatic Search” feature—it will search, redirect you to different websites, and fill out forms for you.
You can watch the AI navigate webpages or switch to another tab while it works in the background. Auto-browsing is possible across multiple tabs, allowing you to multitask. The AI will display its steps in a sidebar as you navigate, making it easier to monitor the process.
Google demonstrated this to journalists, showing how the AI could find a specific product, navigate to its store page, sign in to the buyer’s account (using Google Password Manager), and add it to the cart. The company also suggested using Auto Browse for scheduling appointments, filling out online forms using information from an uploaded PDF, collecting tax documents, comparing apartments listed on sites like Redfin, and much more. I haven’t had a chance to test this in practice yet, so I can’t say how well it will handle these tasks, although in a controlled demo, it seemed quite fast.
Do you really trust artificial intelligence to find information for you?
My biggest concern with the Auto Browse feature is the suspicious sites and permissions it requires, though Google has assured me this is planned. Auto Browse must obtain permission before accessing your Google password manager, and if it encounters a link that its AI deems suspicious, it will presumably use Chrome’s existing unsafe browsing protection to redirect you to another page. A Google spokesperson told me this is “as secure as it can be,” though I’d probably recommend keeping an eye on it, at least for the first few requests.
This feature currently has one limitation: although it can be opened in multiple tabs simultaneously, tabs running AutoView cannot interact with each other. This means each instance of AutoView is isolated, but this may change in the future.
Personally, I don’t think I’ll use it often, especially for sensitive tasks like “tax filing,” but auto-filling a simple form sounds convenient. Google stated that Auto Browse will pause and ask you to take over important steps in any tasks that require it, such as purchasing an item or submitting a form. It won’t (or shouldn’t) perform this last step for you, giving you the opportunity to review your work. In this sense, it’s similar to the existing in-app purchases in Gemini.
Existing and future features
The Gemini Chrome app can leverage most of Gemini’s existing features, allowing it to connect to apps like Gmail or access your chat history with the bot. But there’s one important feature planned for “the coming months.” Gemini recently released a beta version of ” Personal Intelligence ” for paid users, allowing the AI to view all your past conversations and connected apps without you having to tell it where to look. Essentially, this is an extension of the existing app connection and history features, using a logical reasoning model. For example, you could ask it to help you find new tires for your car, and it would automatically understand to look through your Gmail and Google Photos to find out what model your car is and when you last bought tires.
This feature is still in development, but its availability means Google is moving quickly to ensure consistency across all methods of accessing Gemini. All other features mentioned in this article are either already available or are currently being rolled out.