Google’s Gemini automates multi-step tasks on Android

Google introduced Gemini Intelligence for Android, an on-device artificial intelligence feature designed to automate multi-step tasks across apps and personalize device interfaces. The company plans to start the rollout this summer on Samsung Galaxy S26 and Google Pixel 10 phones, with broader availability to other Android-powered devices later in the year.

Gemini Intelligence can carry out sequences of actions that currently require users to switch between apps. Google presented examples such as long-pressing the power button over a grocery list in a notes app and asking Gemini to create a delivery shopping cart with all the items, eliminating manual copying and pasting between services. The feature can also locate a class syllabus in email and add required books to an online cart.

The software includes AI-assisted browsing in Chrome to help complete tasks found online, expanded autofill that pulls information from connected apps, and a multilingual voice-cleanup tool called Rambler that converts spoken input into concise messages. Users will be able to create custom Android widgets with natural language prompts. Google also showed a redesigned interface based on Material 3 Expressive aimed at reducing visual clutter.

Google said Gemini Intelligence will act only after a user issues a command and will stop once the assigned task is finished. Final confirmations are required before purchases or other data changes are completed, and Google emphasized control and privacy in the feature design.

The initial rollout covers the Galaxy S26 and Pixel 10 this summer. The company expects to extend Gemini Intelligence to watches, cars, glasses and laptops that are tied to the Android ecosystem later in the year. At the same time, Google introduced the Googlebook, described as the first laptop built for Gemini Intelligence, but the company did not provide a timeline or say whether the Googlebook will replace the Chromebook.

The announcement follows scrutiny of delayed or undelivered AI features from other smartphone makers. In a recent court settlement, one competitor agreed to pay $250 million over claims that promised AI features were delayed or not delivered on new phones. That competitor has said it will use Gemini to help power parts of its voice assistant.

Google pointed to its years of work on Gemini models, Android integrations and AI infrastructure as preparing the company to run agent-like functions directly on consumer devices. In a statement, the company wrote: “Gemini will navigate tasks for you-whether it’s snagging a front-row bike for your spin class or finding your class syllabus in Gmail then putting the books you need in your cart. Gemini handles the logistics while you stay in the moment.”