Android Studio Agentic AI & Journeys: A New Feature

Android Studio Receives Significant AI Enhancements

Android Studio, the primary integrated development environment (IDE) utilized by developers creating applications for the Android platform, is set to incorporate artificial intelligence capabilities. This announcement was made by Google during its annual developer conference, Google I/O 2025, held on Tuesday.

New AI-Powered Features

Alongside the introduction of the latest Gemini 2.5 Pro model, Android Studio is gaining a novel “agentic AI” functionality known as Journeys. Furthermore, an “Agent Mode” is slated for release, designed to manage more intricate development processes.

Leveraging Gemini, Journeys will empower developers to validate their applications by articulating user journeys and expected outcomes in plain language. Gemini will then autonomously execute these tests.

Google explains that this feature aims to streamline app testing, reducing the need for extensive coding. While currently experimental, the ultimate objective is to accelerate the delivery of high-quality code while minimizing the time spent on testing, validation, and issue reproduction.

Testing can be conducted on either physical Android devices or virtual emulators, with the results displayed directly within the IDE, according to Google.

Agent Mode and Expanded AI Capabilities

An upcoming addition is Agent Mode, an autonomous AI feature also powered by Gemini. This will enable developers to utilize a range of tools to tackle complex, multi-step development tasks. For example, when integrating a new API, the agent can formulate an execution plan, including dependency additions, file modifications, and bug fixes.

Further AI enhancements coming to Android Studio include an improved “crash insights” feature within the App Quality Insights panel, powered by Gemini. This enhancement will leverage AI to pinpoint the source code elements potentially causing application crashes and suggest resolutions.

Studio Labs and Cloud Access

Developers can now explore Google’s experimental AI features through a new “Studio Labs” menu located within Android Studio’s Settings. This option will be available in stable releases, beginning with the version codenamed Narwhal.

The public preview of Android Studio Cloud is also now available. Accessible via Firebase Studio, this service streams a Linux machine running Android Studio to a web browser, facilitating Android development from any location with internet access.

Automated Updates and Code Generation

A Version Upgrade Agent, integrated with Gemini in Android Studio, will soon automate dependency upgrades. Gemini will also assist in automatically generating Jetpack Compose preview code.

Furthermore, developers will be able to transform UI code within the Compose Preview environment using natural language. The ability to attach image files, such as UI mockups or screenshots, to AI prompts, and to include project files as context in Gemini chats, is also being introduced.

A new “Rules in Gemini” feature will allow developers to establish preferred coding styles or output formats.

Enterprise Solutions and Additional Updates

Google is also launching an enterprise-grade version of its AI-powered Android Studio, Gemini in Android Studio for businesses. This allows teams to deploy AI while maintaining data security through subscriptions to Gemini Code Assist in Standard or Enterprise editions.

Other updates include resizable previews in Compose Preview, navigation enhancements, an embedded Android XR emulator that launches in an embedded state by default, and improvements to Backup and Restore and Backup and Sync. Android’s Kotlin Multiplatform will also receive minor enhancements.

Finally, Google will provide developers with tools and early warnings to prepare for Android’s upcoming 16KB page sizes – a fundamental change to Android’s architecture – and facilitate testing applications in the new environment.