Google brings agentic AI and custom widgets to Android

On Tuesday, Google announced a suite of new Gemini Intelligence-branded AI features at its Android Show: I/O Edition event. The announcement marks a push toward agentic AI—which refers to AI capable of performing tasks across apps—by enabling the Gemini assistant to handle complex, multi-step processes. For example, the assistant will be able to copy a grocery list from a notes application and add those items directly to a cart in a shopping app. To initiate this process, users will press the device’s power button and describe the task, while the content displayed on the screen provides context for the assistant. Google noted that Gemini will wait for explicit user confirmation before completing checkouts.

Beyond task automation, the update expands Gemini’s capabilities within other applications. In late June, Android devices will receive Gemini in Chrome, an AI feature designed to help users summarize content or ask questions about web pages, similar to how the tool operates on desktop browsers. The update also brings an auto-browse feature to Android. First introduced experimentally in January, this capability allows Gemini to browse the web and complete tasks such as booking an appointment. Additionally, Gemini will be able to fill out forms on behalf of users by drawing details from Personal Intelligence, which is Google’s feature for learning user details. Google stated that this form-filling feature is opt-in, allowing users to disable it in their settings at any time.

The update also introduces a “vibe-coding” feature—the practice of building apps or widgets via natural language—allowing users to generate custom Android widgets by describing them. For instance, a user can build a meal-planning widget by entering a query such as “Suggest three high-protein meal prep recipes every week.” Google noted that these generated features will follow its Material 3 expressive design language.

These AI-powered tools are scheduled to roll out in phases. Google stated that the features will first arrive on Samsung Galaxy and Google Pixel devices this summer. A broader rollout to other Android devices is scheduled for later this year.

Why it matters

Google is integrating agentic AI capabilities across its Android ecosystem, allowing users to automate multi-step tasks and generate custom widgets via natural language. By embedding these features directly into the operating system, the company is attempting to make AI assistants more action-oriented rather than purely informational.