Google Chrome adds Gemini 3-powered auto browse for paying desktop users

Google announced a new Chrome feature called 'auto browse' powered by its Gemini 3 model in a blog post; the feature is available for Chrome on desktop and is currently rolling out to paying Google AI Pro and AI Ultra subscribers. Auto browse brings an agentic experience to Chrome via Gemini: the assistant appears on a right-hand sidebar, allowing users to chat and interact directly with websites.

Gemini can click, input text, and scroll through pages while carrying out user-requested actions, and it replies 'Task started' when a task begins. Google provided an example in which the AI agent is asked to shop for supplies on Etsy: the user specifies the website, the party theme, and a budget; Gemini searches the site, adds relevant items to the cart, and then informs the user that the checkout cart is ready to review.

Gemini will also look up and add discount codes and, if given permission, access Google Password Manager to log in to the user’s account. A 'Take over task' button hovers at the top of the browser window so a user can intervene, "just in case Gemini goes rogue," the blog post says.

Google said it is continuing to add AI features across its products, noting other Chrome additions such as Nano Banana and further AI Mode integration into Google Search's AI Overviews. Auto browse remains limited to Google AI Pro and AI Ultra subscribers on desktop.

google chrome auto browse, gemini 3, auto browse feature, gemini agentic experience, chrome right-hand sidebar, etsy shopping automation, google ai pro subscription, google ai ultra subscribers, google password manager access, take over task button, nano banana chrome, ai overviews integration