Google's Gemini Spark Now Controls Your Real Chrome Browser to Run Web Tasks
Gemini Spark now controls your real Chrome browser to handle web tasks, and rolls out to AI Pro subscribers in 160+ countries
- Chrome integration: Gemini Spark can now control your real desktop Chrome browser -- not a remote sandbox -- using your logged-in sessions and saved passwords.
- Global rollout: Spark is now available to Google AI Pro subscribers in 160+ additional countries, down from Ultra-only access since May 2026.
- Entry price drops to $20: Previously restricted to the $249/mo AI Ultra tier, Spark is now bundled with the $20/mo Google AI Pro plan.
- Key use-cases: Scheduling apartment viewings, researching and pre-filling flight bookings, updating online orders -- any multi-step web task you'd rather not do yourself.
- Safety guardrails: Payments and social media posts require manual confirmation; Google uses multi-layer prompt injection defenses to prevent malicious websites from hijacking the agent.
- Still excluded: Users in the EEA, UK, Nigeria, and Switzerland cannot access Spark on any plan; the Chrome integration is US-only for now. Official blog
Gemini Spark, Google's 24/7 agentic assistant, just got its most significant capability upgrade yet: it can now take control of your actual Chrome browser to complete multi-step web tasks on your behalf. Paired with that, Google is pushing Spark access to over 160 additional countries for AI Pro subscribers, making this the biggest geographic expansion since the agent launched.
From remote browser to your actual Chrome
Previously, Gemini Spark had access to a remote web browser -- a sandboxed cloud instance. Now it can use desktop Chrome on your device directly, browsing with your logged-in accounts and saved passwords. That's a meaningful shift. A remote browser has no access to your sessions, your cookies, or your credentials. Your local Chrome does.
Spark can use your logged-in accounts and saved passwords to handle tedious web errands, like scheduling viewings for apartments you've saved or researching flight options and starting the booking process. The key word is "starting" -- Spark hands sensitive actions back to you.
How auto browse actually works
Auto browse is triggered by specifying a task when prompting Gemini in Chrome. Google can scroll, click, and enter text on your behalf, with the actions happening on your device while a cloud model is leveraged. Think of it as a remote-controlled cursor running inside a real browser tab, guided by a cloud-hosted Gemini model.
Here's what the experience looks like step by step:
- Gemini confirms "Task started" and opens a new tab badged with a cursor and sparkle icon. Chrome also notes whenever auto browse is active in the top-right corner of your window.