Android Studio, the integrated development environment (IDE) for Android app developers, is getting an AI upgrade, Google announced at its developer conference, Google I/O 2025, on Tuesday.
In addition to the rollout of the latest Gemini 2.5 Pro model, Android Studio is gaining a new “agentic AI” capability called Journeys and will soon introduce an “Agent Mode” for more complex development tasks.
Using Gemini, journeys will allow developers to test their app by describing the actions and assertions in natural language for the user journeys across the app. Gemini will then perform the tests for you.
The feature, explains Google, will let developers test their apps more easily, without having to write extensive code to do so. The company cautioned that this is still experimental, but ultimately, the goal is to increase the speed of shipping high-quality code while reducing the time it takes to test, validate, or reproduce issues.
The tests can run on physical or virtual Android devices, and their results will appear directly in the IDE, Google says.
Another soon-to-arrive addition involves an autonomous AI feature powered by Gemini called Agent Mode. This will allow developers to use various tools to handle more complex, multi-stage development tasks. For instance, if a developer is trying to integrate a new API, the agent may come up with an execution plan that adds the necessary dependencies, edits files, and fixes bugs.
Other AI features coming to Android Studio include a Gemini-powered improvement to the App Quality Insights panel’s “crash insights” feature, which can now use AI to help determine what in an app’s source code may have caused the app to crash, and suggest a fix.
Plus, Google will now allow developers to try out its still experimental AI features through a new “Studio Labs” menu in the Settings menu of Android Studio. This option will be available in stable releases only, starting with the release codenamed Narwhal.
Another experiment now available is the public preview of Android Studio Cloud. Accessed through Firebase Studio, the new service streams a Linux machine running Android Studio to your web browser, enabling Android development anywhere you have access to an internet connection.
A Version Upgrade Agent will soon arrive as part of Gemini in Android Studio to help automate dependency upgrades.
Gemini will also help developers automatically generate Jetpack Compose preview code, transform UI code within the Compose Preview environment using natural language, attach image files (like UI mockups or screenshots) to AI prompts, attach project files as context in chats with Gemini, and set up preferred coding styles or output formats with a new “Rules in Gemini” feature.
The company is also rolling out an enterprise-ready version of its AI-powered Android Studio with the launch of Gemini in Android Studio for businesses, which lets teams deploy AI while keeping data safe when subscribing to Gemini Code Assist in Standard or Enterprise editions, it says.
Other updates include resizable previews in Compose Preview and navigation improvements, an embedded Android XR emulator that launches by default in the embedded state, and upgrades to Backup and Restore and Backup and Sync, among other things. Android’s Kotlin Multiplatform will also see a small handful of improvements.
In addition, Google says it will help developers prepare for Android’s 16KB page sizes — a change to Android’s underlying architecture — with early warnings and tools for testing apps in the new environment.
