Google's personal AI agent Gemini Spark can now work inside your Chrome browser — using your logged-in accounts and saved passwords to handle tedious web errands on your behalf, from scheduling apartment viewings to researching flights and starting the booking process.
Announced in late July and highlighted again this week, the deeper integration builds on Chrome's auto browse, an agentic feature powered by Gemini 3 that performs multi-step chores: comparing hotel and flight costs across dates, filling out forms, collecting tax documents, getting quotes from plumbers, checking bills, managing subscriptions, even renewing a driver's license. With permission, Spark can use Google Password Manager for tasks that require sign-in. Online payments and other sensitive actions pause and ask for human confirmation.
Google is also rolling Gemini Spark out to paid subscribers in over 160 new countries, after first shipping it to AI Ultra subscribers in the U.S. and then to AI Pro users. The agent works in the background 24/7, connects to third-party apps, and can take over tasks mid-flight.
Google says the Chrome integration comes with additional safeguards against prompt injection. The company also introduced the Universal Commerce Protocol (UCP), an open standard for agentic commerce co-developed with Shopify, Etsy, Wayfair and Target, so agents can act on a user's behalf across the web — and sites are increasingly being built with AI agents, not just humans, in mind.
