Google announced on November 17, 2026 that full-stack Android development features are rolling out in Google AI Studio. Its Build agent can now connect generated Kotlin apps to persistent storage, sign-in and Gemini features, with backend setup requiring the user's confirmation.
What Google says is changing
The update adds Cloud Firestore persistence, Firebase Authentication with Google Sign-In, and Gemini integration through Firebase AI Logic and App Check. The browser emulator also gains audio, an HD option and screenshot-to-agent feedback for visual edits.
Google describes publishing to a personal Google Play library for installation on a phone. That is a personal testing route, not evidence that a generated app has cleared a public store release. The announcement describes a rollout; it does not establish that every account already has access.
MintedBrain take: test the complete workflow
The useful question is whether a generated app keeps working beyond its first successful screen. As an illustrative project, a student could build a private revision planner, then check whether a saved task survives closing and reopening the app, whether a second account can see private tasks, and what happens when connectivity drops.
Those are suggested checks, not reported product tests. MintedBrain has not independently tested this release. Treat generated code as a starting point for review, including authorization rules, error handling and service costs, before relying on it for other people's data.
For foundational lessons, explore MintedBrain Academy and the tutorial library.
Discussion
Sign in to comment. Your account must be at least 1 day old.