Gemini Task Automation Is Now on Pixel 10: What It Does and How to Enable It

Last Updated on April 10, 2026 by Alphabet Insider Staff

Google’s AI can now handle multi-step tasks inside apps, and Pixel 10 owners are next in line

Pixel 10 users, hear this! You can now offload daily tasks to your phone’s nifty AI assistant. Gemini AI can now order your lunch, book your Uber, order takeout, and fill your grocery cart without opening a single app. That’s the promise of Gemini task automation, and as of this week, it’s available to Pixel 10 owners in the US.

This isn’t a concept or a demo. It’s live.

What Is Gemini Task Automation?

Gemini task automation lets the AI complete multi – step tasks inside supported apps , running them in a secure virtual window on your phone. What happens in that window is processed in the cloud and it can’t access the the rest of your device.

You can watch Gemini scroll, tap, and type in real time inside that window, or you can keep using your phone normally while it works in the background. When it’s running a task, you get notifications so you can jump in at any point.

The core idea is simple: you tell Gemini what you want, and it handles the tedious tapping for you.

How It Works on Pixel 10

You trigger it the same way you already access Gemini, by holding the power button or using “Hey Google.” Once you give it a task, Gemini walks through it step by step on screen in a virtual window, and you can take control at any time.

On first launch, you agree to “Get tasks done with Gemini.” From there, you’ll see a “Working on this task” card and a “Task in progress” system notification that lets you view progress or stop the task entirely.

Importantly, Gemini will prompt you to open the app to tap the actual buy or order button yourself. You have to explicitly grant permission before Gemini automation activates, and it stops as soon as the task is finished. Google built a deliberate confirmation step into the checkout flow, which is a reasonable safeguard given that the AI is handling real purchases.

What Apps Does It Support Right Now?

The feature currently supports Uber and foodie apps like Uber Eats, Grubhub, Doordash and Starbucks. Gemini can also ask follow-up questions like drink size you want or store location when placing an order.

That’s a narrow list for now, but the apps chosen cover the highest-frequency daily tasks for most people. Wider support is expected as Google refines the system.

How It Got Here

Gemini screen automation debuted at Samsung’s Galaxy Unpacked event in February 2026, where Samsung and Google showcased the feature on the Galaxy S26 series. The Galaxy S26 got it first when the phone launched on March 11.

Google further detailed the feature with the March Pixel Drop, which also marked its rollout to Google’s own devices. The feature is now available across the entire Pixel 10 lineup, including the Pixel 10, Pixel 10 Pro, Pixel 10 Pro XL, and Pixel 10 Pro Fold, running Android 16 QPR3 stable.

Usage Limits Matter

This isn’t unlimited. Free users can make up to five requests per day. Higher tier plans extend the limits: 12 for AI Plus, 20 for AI Pro, and up to 120 for AI Ultra.

In contrast, 5 requests a day is thin if you rely heavily on delivery apps, but for most casual users it’s likely enough to test the feature and build a habit about it.

One Privacy Note Worth Knowing

Google cautions that Gemini takes screenshots during automation sessions, which may include visible app information. These images go to trained reviewers for service improvement when the Keep Activity setting remains enabled.

That’s standard for AI product improvement pipelines, but worth knowing before you enable it.

The Bigger Picture

Gemini task automation is the clearest sign yet that Google’s AI assistant is ever-evolving and is actually doing things. Not just answering questions. It’s a meaningful shift in how people might use their phones, and the Pixel 10 is now at the center of it.

To enable it on a Pixel 10 running Android 16 QPR3, open the Gemini app, go to App Settings, tap Screen Automation, grant the relevant permissions, and agree to the terms. It’s that straightforward.