Google’s Gemini Could Soon Automate Your Phone Tasks on Android
Google is developing a new functionality for its Gemini AI assistant which will allow users to operate Android devices through their virtual assistant. The Google app beta test includes screen automation functionality which enables Gemini to complete tasks such as food ordering and transportation booking without requiring any user input.
Developers found evidence of this feature in the Google app version 17.4 beta which contains internal code strings that demonstrate how Gemini assists users in achieving their objectives within compatible applications. According to these strings users could issue commands to Gemini which would execute their requests for product ordering and ride reservation through the application interface.
The feature receives the internal designation of Get tasks done with Gemini and it will use automated processes to control applications without requiring users to make manual selections. Google will restrict its initial automation solutions to particular applications which they will expand to additional programs in subsequent software updates.
Google has not announced any details about a public release date while testing remains under evaluation. The APK teardown process reveals possible upcoming features which might undergo modifications or become permanently abandoned. The company provides information about its current development projects through these leaks.
To make such automation work, Google appears to be building support directly into the Android operating system, with references linking the feature to Android 16 QPR3, a planned update for Google’s mobile platform. This indicates that the system’s underlying infrastructure may be necessary to let Gemini interact with apps in a controlled way.
The primary goal of screen automation technology uses AI assistants to execute tasks instead of providing user recommendations. An AI system like Gemini has the ability to perform booking tasks and order form completion through its digital assistant functions which extend beyond simple information retrieval. The success of this project will create a major transformation in how users operate their mobile devices which will affect their performance in both simple tasks and complex procedures that require multiple steps to complete.
There are essential requirements for user protection and operational security. The AI system will perform essential functions which involve handling confidential information during its interaction with mobile applications of users. Users must establish complete authority over Gemini’s operational abilities while developers need to establish effective protection methods which prevent automation from resulting in unintentional results and security vulnerabilities.
Industry observers note that Google has been gradually expanding Gemini’s role across its products. Gemini already powers features in search results and deeper integrations with apps like Gmail and Maps. Allowing it to automate tasks on Android devices represents a broader push toward “agentic AI”, where assistants do work for users instead of only responding to questions.
For now, screen automation remains under development. Users who want to try experimental features might need to join the beta channel for the Google app or Android QPR3 when it becomes available. But the groundwork spotted in code shows that Google is serious about giving its AI assistant more proactive capabilities, and that the way people interact with mobile devices could change significantly in the near future.
