Gemini is now equipped with a new update from Google that streamlines multi-step tasks on Android devices. This innovative chatbot can manage tasks like placing orders through food delivery services or booking rides by navigating various third-party applications independently.
Google is said to have developed a secure virtual window that changes Android’s operating system from a passive tool into an active AI assistant.
This feature is being launched as an early preview and will soon be accessible as a beta option within the Gemini app for select Android devices, including the Pixel 10, Pixel 10 Pro, and Samsung Galaxy S26 series. Initially, Gemini will be compatible with chosen applications in categories such as food, grocery, and ridesharing.
How Will Gemini Automations Work?
The multi-step automation functionality of Gemini will serve as a “personal assistant” on Android devices, guiding users through third-party applications to accomplish specific tasks. To start, users need to activate Gemini by issuing a command like “Order my usual pepperoni pizza from DoorDash.” Gemini will evaluate the app, review the order history, and navigate to the checkout page for the final step.
Google has ensured that users will “monitor Gemini’s progress live via notifications, allowing them to view, intervene, or halt the task as needed.” The firm asserts that automation occurs within a Secure Virtual Window, allowing it to only access and interact with the particular application needed for the task. This development aims to save users both time and effort when handling repetitive tasks.
The Gemini multi-step task feature is currently being rolled out for users in the U.S. and Korea. With this introduction, Google is also entering the competitive AI space, joining competitors like Anthropic and OpenAI, who are advancing their enterprise AI agents. Nevertheless, Google’s initiative brings this level of autonomous execution directly into users’ hands.