Why browse the web when an AI can do it for you? Google is amping up the AI on its Chrome browser with an “auto browse” mode that can automatically complete tasks for you, like buying things online or scheduling an appointment.
Rivals such as Microsoft, OpenAI, and Perplexity offer similar “agentic” AI modes that can take over the browser and automatically get to work. Google has already been integrating its Gemini assistant into Chrome; it even embedded a dedicated button for the chatbot. Now the company wants the AI to handle complex, multi-step workflows.
“Our testers have used it for all sorts of things: scheduling appointments, filling out tedious online forms, collecting their tax documents, getting quotes for plumbers and electricians, checking if their bills are paid, filing expense reports, managing their subscriptions, and speeding up renewing their driving licenses—a ton of time saved,” Google says in a blog post.
The company posted a demo in which a user activates auto browse mode by asking Gemini to buy decorations from Etsy to match an image from their Gmail inbox. The AI navigates to Etsy through the browser and successfully adds the items to the user’s cart.
The AI is even able to personalize one of decorations by adding the text “Y2K Party,” before offering the user a summary of the listed items and telling them they can complete the checkout process. In another demo, auto browse mode is able to take data from a document and use it to fill out an online form.
That said, auto browse mode isn’t perfect. Google also demoed the feature to journalists by asking the feature to navigate to a website and re-order a previously purchased jacket in the same size and to look for potential discounts. The AI first asked permission to access the user’s login information saved in Chrome. But the AI seemed to be slow in processing the full request, an issue we’ve encountered with other AI browser integrations.
Get Our Best Stories!
Your Daily Dose of Our Top Tech News
By clicking Sign Me Up, you confirm you are 16+ and agree to our Terms of Use and Privacy
Policy.
Thanks for signing up!
Your subscription has been confirmed. Keep an eye on your inbox!
For sensitive functions, such as completing a purchase or posting on social media, the auto browse has been designed to hand over the final step to the user. It’s a trade-off that can prevent errors, but also undercut the convenience factor.
(Credit: Google)
We’ll have to try it ourselves. But we hope it’s an improvement over Project Mariner, Google’s experimental AI browser. In our testing last year, Mariner sometimes struggled to understand commands or successfully interact with third-party sites or overcome CAPTCHA pages.
Recommended by Our Editors
For now, Google is merely “previewing” the auto browse mode for Chrome. It will be available first in the US to paid subscribers of Google AI Pro and Ultra, which cost $20 and $250 per month, respectively. The other catch is how auto browse mode features a cap on the number of user requests per day. Pro subscribers can complete 20 requests while Ultra offers 200. Users can run multiple tasks at the same time.
The company is likely holding off on a mainstream release to work on potential kinks. A major concern facing AI browser is how they can hallucinate the wrong information and fall for “prompt injection attacks,” which can trick the AI into performing a malicious action. However, Google has already been working on “layered defenses” to guard against such threats.
Although it remains unclear if or when auto browse mode will arrive to free users, Google is laying the groundwork for the feature and Gemini to be able to connect with third-party platforms. The company noted Chrome “will support Google’s Universal Commerce Protocol (UCP), a new open standard for agentic commerce co-developed with industry leaders including Etsy, Shopify, Target, and Wayfair. This new open standard ensures that AI agents can understand and interact with different websites seamlessly.”
About Our Expert
Michael Kan
Senior Reporter
Experience
I’ve been a journalist for over 15 years. I got my start as a schools and cities reporter in Kansas City and joined PCMag in 2017, where I cover satellite internet services, cybersecurity, PC hardware, and more. I’m currently based in San Francisco, but previously spent over five years in China, covering the country’s technology sector.
Since 2020, I’ve covered the launch and explosive growth of SpaceX’s Starlink satellite internet service, writing 600+ stories on availability and feature launches, but also the regulatory battles over the expansion of satellite constellations, fights with rival providers like AST SpaceMobile and Amazon, and the effort to expand into satellite-based mobile service. I’ve combed through FCC filings for the latest news and driven to remote corners of California to test Starlink’s cellular service.
I also cover cyber threats, from ransomware gangs to the emergence of AI-based malware. Earlier this year, the FTC forced Avast to pay consumers $16.5 million for secretly harvesting and selling their personal information to third-party clients, as revealed in my joint investigation with Motherboard.
I also cover the PC graphics card market. Pandemic-era shortages led me to camp out in front of a Best Buy to get an RTX 3000. I’m now following how President Trump’s tariffs will affect the industry. I’m always eager to learn more, so please jump in the comments with feedback and send me tips.
Read Full Bio
