With this feature, Gemini will learn to control the Chrome browser. It will help manage tedious routine tasks

Google is apparently working on expanding the capabilities of its Gemini artificial intelligence. According to findings by TestingCatalog, a new Auto Browse tool has appeared in the internal interface, which should allow AI to autonomously browse the web and control the Chrome browser. The feature is not yet publicly available, but its presence suggests it could appear relatively soon.

What will the feature be good for?

The Auto Browse tool should allow Gemini to manage tabs, browse web pages, and perform tasks directly in the Chrome browser instead of the user. Available information suggests that the feature will be integrated into the Chrome sidebar, from where it can be conveniently controlled.

Google is working on a new tool for Gemini called "Auto Browse". Auto Browse will Chrome into a proper agentic browser, where Gemini will serve as a control interface.

This addition is a continuation of agentic functionalities announced in September. pic.twitter.com/uSfS8VkuNS

— TestingCatalog News 🗞 (@testingcatalog) January 13, 2026

Google will respond to the competition with this move. The popular AI search engine Perplexity operates a so-called agent browser Comet, which attracts users with its ability to perform routine tasks on their behalf. The Atlas browser from OpenAI, the organization behind the generative tool ChatGPT, is also built on de facto the same idea.

For Subscribers Only

Based on the information available to it, TestingCatalog anticipates that the Auto Browse feature will be exclusive to the Gemini Ultra plan. It will thus primarily target professionals and demanding users who want to automate research or routine web tasks.

Google already tested the first agentic features for Chrome last September, but only to a limited extent for US users. A wider rollout of Auto Browse could come soon.

Would you use an AI assistant to control your browser?

Source: TestingCatalog