Google LLC today introduced major updates to Gemini in Chrome, the company’s flagship artificial assistant in its major web browser, that will allow users to access the experience in new ways.

With the new updates, users will be able to multitask on the web with a new side panel, generate images directly in the browser, automate multistep processes using “auto browse” and, in the coming months, add personal intelligence to the assistant.

Let Chrome and Gemini do the work for you

Google is welcoming users to a brand-new era with “auto browse,” which is rolling out in preview for subscribers to Gemini Pro and Ultra.

Essentially, users may now delegate tasks to Chrome by asking it to do tasks for them on the web, including filling out forms, logging into websites, shopping and even making purchases.

During a demonstration, Parisa Tabriz, vice president of Chrome at Google, used auto browse to reorder a gray jacket. This included logging in — after requesting her permission and getting her credentials from Google Password Manager — going through the process of finding the jacket in the store, putting it in the cart and completing the purchase.

She even tabbed away as auto browse operated to continue talking about a different part of the demo while the agent worked in the background.

Auto browse can help optimize vacation planning by doing mundane work, such as researching hotel and flight costs across dates to find budget-friendly times for travel. It can also help schedule appointments, fill out tedious online forms, collect tax documents, get quotes for plumbers and electricians, check if bills are paid, fill expense reports, manage subscriptions and more.

Gemini in a sidebar

Google brought Gemini to Chrome in late 2025, providing access to its AI assistant to users on MacOS, Windows and Chromebook Plus. At the time, when Gemini was activated, it opened in its own little pocket window that users could drag around on the screen, which had awareness of the open tab.

Starting today, users can attach it to the side of the browser window. If users open another tab, they can also have Gemini attach to those and only use that one as context.

However, if users want Gemini to consolidate across multiple tabs – say if they are comparing products, services or planning a trip, they can click open multiple windows and the Gemini tab will add all the tabs to context. The users can then chat with Gemini in the side panel about their tabs all at once.

According to Google, test users can then keep their primary tab open and use the side tab to keep a single conversation thread with Gemini and compare options.

Transform images on the web

By bringing the creative power of Nano Banana directly into Chrome, Gemini now allows users to transform images into whatever they like directly in their web browser without needing to download and re-upload them.

Users need only type a prompt into the side panel while an image is displayed in the main window and tell Gemini what they want and the AI assistant will do the rest. For example, they could have a photo of their bedroom and ask Gemini to look at another tab with a lamp that they’re interested in, to have it add it so they can see what it might look like next to their bed.

Chrome becomes more personal with AI

Google also said in the coming months it will roll out a more personalized experience by bringing Personal Intelligence, a capability it recently brought to the Gemini app, to Chrome. With this capability, Chrome will remember context from past conversations so users get uniquely tailored answers from web searches that fit their personality and needs.

The company also stressed that this is an opt-in experience and users can choose which apps to connect or disconnect at any time.

Images: Google