Google’s New Chrome ‘Auto Browse’ Agent Attempts to Roam the Web Without You (www.wired.com)

Google has launched a new "Auto Browse" feature for its Chrome browser, utilizing the Gemini 3 generative AI model to automate online tasks such as booking flights, finding apartments, and managing expenses. Currently available only to subscribers of Google’s AI Pro and Ultra plans in the U.S., Auto Browse exemplifies Google's ongoing commitment to integrating AI into web browsing. This release follows last year's introduction of the "Gemini in Chrome" mode, which provided users with AI-assisted answers and synthesis of web content. The significance of Auto Browse lies in its potential to redefine user interaction with web browsers, aligning with a broader trend in Silicon Valley toward more AI-driven experiences. The feature operates by autonomously navigating web pages and completing tasks based on user prompts, although it maintains a level of necessary oversight for sensitive actions. While this automation presents exciting possibilities for efficiency, it also raises important security concerns, particularly regarding vulnerabilities to prompt injection attacks on untrusted sites. As Google gradually rolls out Auto Browse, it signals a shift towards a future where AI plays an increasingly integral role in everyday digital activities.