Gemini Screen Automation in India: How to Set Up Android App Tasks

Gemini screen automation is rolling out in India on supported Android devices. The beta feature lets the Gemini mobile app open selected apps and handle steps such as adding groceries to a cart or preparing a ride request. Google’s November Pixel Drop announced the expansion to India, Japan and the UK. The workflow still asks you to complete final steps such as confirming a ride or placing an order.

Start with the Gemini Apps Help eligibility list. It names Pixel 10, 10 Pro and 10 Pro XL, the Pixel 11 family, and selected Samsung devices including the Galaxy S26 series. The account must be a personal Google Account and the user must be at least 18. The beta supports selected apps in English, Japanese and Korean. In India, try the first task in English if you want to check whether it is available on your device.

Start with a task you can easily review

Open the Gemini app on Android and describe what you want done. For example, ask it to add milk and eggs to a grocery cart in a supported app. Include useful constraints, such as which items you want and the maximum quantity. For a ride, specify the pickup and destination before checking the details Gemini prepares. A short request with a clear end point makes the plan easier to review.

Gemini may ask follow-up questions or use a connected app if that route can handle the request. If screen automation is needed and this is your first use, follow the on-screen prompt to grant the Google app permission for the relevant app. Screen automation lets Gemini read and interact with that app’s screen. The permission applies per app, so a successful first task does not give every installed app the same access.

Check any plan that Gemini shows before agreeing to it. If the plan names the wrong address or item, reply with the change before moving on. Gemini can work in the background. Use the chat or live update notification to return to the task and see how far it has progressed.

Confirm orders and know how to stop a task

Google says to follow the instructions in the chat for any steps that need your input, including the final confirmation. Compare the app’s address, quantities and total before confirming an order. If you stop or leave the process incomplete, check the app directly. Google says items that Gemini added may remain in the cart after an unfinished task.

To stop the automation, use the stop control in Gemini or the stop task action in the live update notification. The help page says that taking control of the task is currently unavailable on Pixel 10 and Galaxy S26 devices. If you need to enter sensitive information on one of those phones, stop the task and open the service’s own app to finish. Keep passwords and payment details out of the Gemini chat.

Where to review app permissions

In Gemini, open the menu, select your profile picture or initial, then open Settings and look for Screen automation. Choose the relevant app to review its access. If the control is missing, check the device, language and account requirements first. Google’s rollout announcement says November updates will continue over several weeks, so access can arrive later on a device that meets the requirements.

When the feature is available, use the first task to learn which steps Gemini can prepare and which are still yours to complete. The most useful check is in the service app itself, where you can see the final cart or ride request before approving it.