Chrome’s New ‘Automatic Browsing’ Agent Attempts to Surf the Web Without You

Google released ia Chrome’s new “Auto Browse” feature on Wednesday. The tool, powered by Google’s proprietary Gemini 3 AI model, is an AI agent designed to take over your Chrome browser to help complete online tasks such as booking flights, finding apartments, and filling expenses.

The release of Auto Browse is part of Google’s continued integration of AI features into Chrome. Last year, Google rolled out the “Gemini in Chrome” mode to answer questions about what’s on web pages and aggregate information from multiple open tabs.

Browse Automatically, which users can access by launching the Gemini sidebar in Chrome, will be available today only in the US to subscribers of Google’s monthly AI Pro and AI Ultra plans. It is not clear when Auto Browse will be available to premium users in additional countries.

Google is going all out with Silicon Valley’s vision of the future of web browsing, which involves a lot more AI and a lot less of you. Whether it’s a browser designed from the ground up around generative AI, like OpenAI’s Atlas, or one retrofitted with new AI-based tools, like Google’s Chrome, almost every option available to consumers now has some level of AI baked in. (The Vivaldi browser is different for users who want to avoid AI-powered web browsing.)

In a pre-launch demo, Charmaine D’Silva, Chrome’s director of product management, showed me an example of Auto Browse for online shopping. “Instead of having to remember where I bought something and try to reorder something,” he said, “now I can send to AutoBrowsing within Gemini so I can go ahead and shop for jackets.” By typing a message into the Gemini sidebar in Chrome, D’Silva asked the bot to reorder a jacket he bought last year, and get a discount coupon code before making the purchase.

When activated, Auto Browse takes over Chrome and makes ghost clicks on its tab while trying to complete a given task. “Use Gemini with care and control if necessary,” reads the disclaimer on the demo version. “You are responsible for Gemini’s actions during work.” Even if you send it to the digital wilderness, Google still sees you as responsible for what its bot does online while following your requests.

For now, automation only goes so far. Activities that Google considers more sensitive, such as posting on social media and swiping your credit card, still require less user supervision. In these cases, the Chrome bot will log the steps it took to get that far and ask the user if they’d like to continue.

Anyone interested in trying out AutoBrowsing should strongly consider the security implications of this type of automation. Despite Google’s efforts to make it safer to use, Auto Browse and similar AI-based tools are still vulnerable to being fooled by rapid injection attacks when visiting malicious websites, tricking the bot into acting in a way the original user did not intend.

I’ll be testing Auto Browse this week to get a feel for its initial strengths, weaknesses, and what this tool means for average Chrome users. In general, I don’t trust AI tools designed to streamline your life and sweep all your digital activities. Bots are always overcrowded, and I’ve found them to be unreliable. Even so, Google insists on reshaping the web browsing experience with AI. Given Google’s track record of pushing out new features gradually, you should expect Auto Browse to become more widespread in the near future.