Google AI Studio Now Builds Full-Stack Android Apps From a Prompt and Ships Them to Google Play

Google AI Studio can now build full-stack Android apps from a single prompt. The Build agent sets up Firestore, Google Sign-In and secure Gemini API calls through Firebase, and apps can be published to a personal Google Play library. Custom domains arrive for web apps.

Christopher Ort
By Christopher Ort · 3 min read
Share
Hands holding a phone beside a laptop on a wooden desk.
Image: AI-generated, i10X
On this page
  1. Quick Take
  2. In plain English
  3. What happened
  4. Background
  5. How it works
  6. Why it matters
  7. What's still open
  8. Sources
  9. i10x Perspective

Google AI Studio can now build full-stack Android apps from a single prompt and publish them to a personal Google Play library, Google announced on 17 November 2026. The Build agent, running on Gemini 4 Carbon and Hydrogen, sets up the database, sign-in and secure Gemini API calls on its own.

i10X Quick Take

Quick Take

  1. What: native Kotlin and Jetpack Compose Android apps generated in Google AI Studio Build.
  2. Backend: automatic Cloud Firestore database with Security Rules and Google Sign-In through Firebase.
  3. Security: Gemini calls go through Firebase AI Logic and App Check, so no API key sits inside the app file.
  4. Publishing: apps ship to a personal Google Play library with a standard Google account.
  5. Also new: custom domains for published web apps on paid Google Cloud projects.

In plain English

Google AI Studio is Google’s free web tool for trying Gemini and “vibe coding” apps by describing them in words. Until now it mostly built web apps. Now you can ask for a phone app, test it in an Android emulator in the browser and install it on your own phone through Google Play. The tool also handles the parts beginners usually get stuck on, such as user accounts and saving data.

What happened

In a Google Developers Blog post, AI Studio product manager Mike Taylor-Cai wrote that the tool lets developers “go from a single prompt to a production-ready Kotlin Android app complete with data persistence, user sign-in, and AI features.” The features are rolling out today. Developers select the Android chip at the top of the Build workspace to start.

Background

Google connected Firebase to AI Studio for web apps earlier this year. The Firebase team says the Android version needs no installed software, no manual setup and not even a credit card. The launch builds on Google’s recent push with Gemini 4 Carbon and other developer tools.

How it works

  • Emulator: a faster, high-definition Android emulator with audio output runs in the browser.
  • Visual edits: one click captures a frame from the running emulator and attaches it to the prompt to guide UI changes.
  • Data: the agent provisions Firestore, configures the Firebase SDKs and writes Security Rules for the app’s data model, with offline caching that syncs later.
  • Accounts: Firebase Authentication with Google Sign-In, plus per-user rules so people only see their own data.
  • Tooling: custom fonts for Compose UIs and the official Android CLI for project scaffolding and builds.
  • Domains: on paid Google Cloud projects, users can buy and connect a domain from the Publish panel. AI Studio registers it in Cloud Domains, sets up Cloud DNS and creates a managed SSL certificate. Free ai.studio subdomains remain available.

Why it matters

For hobbyists and founders, a working Android prototype with logins and storage is now a prompt away.

For developers, keeping API keys out of the app file closes a common security hole in AI-generated mobile apps.

For Google, tying AI Studio to Firebase, Play and Google Cloud pulls new builders into its paid stack.

What’s still open

  • How well generated apps hold up beyond simple prototypes.
  • Which regions get personal Play publishing.
  • When AI Studio supports transferring existing domains, which Google says it plans to add.

Sources

  • Google Developers Blog, “Build full-stack Android apps with Gemini in Google AI Studio,” Mike Taylor-Cai, 17 November 2026
  • Firebase Blog, “Build full-stack Android apps with AI features using Google AI Studio and Firebase,” 17 November 2026
  • Google AI Studio documentation, “Publish your app in Google AI Studio”

i10x Perspective

Prompt-to-app tools have mostly stopped at the browser. Putting a real Android app on a real phone in minutes is a meaningful step, and it competes directly with Replit, Lovable and Bolt. More on our Google company page and in the Coding section.

Christopher Ort

Christopher Ort

Christopher Ort handles SEO, AI agent workflows and PR & partnerships at i10x.ai, the all-in-one AI workspace built by i10X Global Pte. Ltd. in Singapore. He is a graduate of Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU). He is the author of “The Style Penalty”, a study published on SSRN that analyses 1,576 evaluation decisions by four frontier AI models to show how AI-written CVs fare in automated hiring screens; the research has been covered by heise online and t3n.

Timeline

What happened before

  1. Release

    Gemini 4 Argon

    1.05M context · $2 / $10 per 1M tokens

    Google's most capable model for complex coding, agentic, math and science tasks.

    ai.google.dev (source, opens in a new tab)
  2. Release

    Gemini 3.8 Flash

    1.05M context · $0.75 / $3.75 per 1M tokens

    ai.google.dev (source, opens in a new tab)
  3. Release

    Gemini 3.7 Flash

    1.05M context · $0.75 / $3.75 per 1M tokens · Deprecated 8 Oct 2026 (API routes to Gemini 3.8 Flash)

    ai.google.dev (source, opens in a new tab)
  4. Release

    Gemini 3.6 Flash & 3.5 Flash-Lite

    1.05M context · $0.75 / $3.75 per 1M tokens

    ai.google.dev (source, opens in a new tab)
  5. Release

    Gemini 3.5 Flash

    1.05M context · $1.5 / $9 per 1M tokens · Deprecated 8 Oct 2026 (API routes to Gemini 3.6 Flash)

    Launched at I/O 2026 as the first Gemini 3.5 model; Google later dropped the planned 3.5 Pro.

    blog.google (source, opens in a new tab)
Full Gemini timeline ›

Editor's picks

  1. CVE-2026-132121: Any website can hijack a local Supergateway and run your MCP tools
  2. CVE-2026-132122: agentic-radar scans can leak your OpenAI API key to a malicious repository
  3. CVE-2026-119220: Dokan MCP endpoint regex flaw lets attackers bypass WooCommerce permissions
  4. Reflection AI Gates Laser, Its Autonomous Penetration Testing Agent, Behind a Trusted Access Program
  5. BrightAI Raises $97 Million to Replace Scheduled Infrastructure Inspections With Physical AI