Forgot your password?
Close
typodupeerror
Android AI Google Build

Google's AI Studio Now Lets Anyone Build Android Apps In Minutes (techcrunch.com) 39

Posted by BeauHD from the weeks-to-minutes dept.
An anonymous reader quotes a report from TechCrunch: The AI coding boom is now coming directly for Android app development. On Tuesday at Google IO 2026, the company announced new native Android app creation capabilities in its web-based Google AI Studio, shrinking a process that takes weeks of setup and coding down to minutes. The company also said that consumers will be able to use Gemini AI to find the apps they need, both on the Play Store and the web, expanding opportunities for developers to have their apps discovered.

Google says the new capabilities could make sense for anyone from a seasoned developer looking to prototype a new app quickly to a first-time creator. [...] The apps are built with the Kotlin programming language using Google's Jetpack Compose toolkit and with support integration with hardware sensors like GPS, Bluetooth, and NFC, the company says. However, the resulting creations, for now, are only meant to be used personally, as publishing for family and friends is still on the roadmap. The company suggests the technology could be used for the creation of personal utilities and simple social apps, hardware-enabled experiences, or AI-powered experiences. Google is also adding an "Ask Play" AI overlay to the Play Store that lets users discover apps through natural-language conversations. "Perhaps more importantly, apps will begin to be surfaced with users' conversations with Google's Gemini virtual assistant, exposing developers' apps to millions of users," adds TechCrunch.

Google's AI Studio Now Lets Anyone Build Android Apps In Minutes More | Reply

Google's AI Studio Now Lets Anyone Build Android Apps In Minutes

Comments Filter:

  • Where is the shovelware? Where's the killer app? (Score:5, Interesting)

    by MIPSPro ( 10156657 ) on Wednesday May 20, 2026 @11:21AM (#66152687)
    Where is the Shovelware? [substack.com]. Surely they could vibe code up a couple dozen killer apps to prove their point. Where are they? Show me how many of the top 10 Android apps were built this way. How many of the top 100? It's like "Loveable" web-sites. They are simple, they suck, they leak data like sieve. Nobody wants shitware, folks, they want real working useful applications or fun/creative/attractive games. So far I've seen exactly ZERO made by AI.

    • Re: (Score:2)

      by dfghjk ( 711126 )

      "Nobody wants shitware..."

      If only that were true. No consumers want shitware but most tolerate it. The industry LOVES shitware, the lower the bar the easier the profit.

      • Re: (Score:2, Insightful)

        by shanen ( 462549 )

        So what's the solution? How about if the google revealed the business model of each app to give us the information to recognize the shovelware? (I'm interpreting that term based on memories of a couple of days actually shoveling horse manure back when I was in the service... So long ago that I can't remember the details, but I think the first day we were piling it in one place and the second day we had to get it back in the truck because they decided it was the wrong place...)

        I know solutions are unfashiona

      • Well, okay, I guess you have some point. The weasel class definitely would love to vibe code shitware then shove it down our throats. I expect that when I use, say, my insurance company's crappy app. However, I feel like these vibe coded apps really just contribute to the same perception I have: AI just produces low-quality shitware and even that just barely.
        • I personally find the bigger problem with slopware is that it buries the genuinely good software out there. Inevitably whenever someone does produce some software of value people will come along and clone it maybe add features. That used to require effort and a modicum of talent. Now there's 50 people all trying the same thing for everything that could generate a penny of revenue. Users try a few claiming to solve their issue and give up. Sure I got old my craft is useless whatever slopware devs tell themse

      • While I agree that many in the industry want the cheapest and the fastest to build regardless of quality, my question is about the demand side.

        Who wants these apps? What is is that they do that someone is willing to pay for? How does that address the cost of the other inputs that make apps worth enough money or other rewards that someone wants to maintain them?

        We are 18 years out from the launch of iPhone App store, and even though humans are far slower than AI in building apps, after nearly two decades I d

    • Because I am tired of shitware, and because I don't have time to code Android apps from scratch, this will let me make my own Android apps for my own use, thus avoiding all the shitware.

      • It's the Internet. I can't tell if you're a dog or if you're teasing. If you're serious the problem you'll run into is that vibe coded apps are very often shitware. In fact, nearly always. Only blended hybrid apps seem to actually function and those require a real programmer, not a viber.

    • Re: (Score:2)

      by allo ( 1728082 )

      "So far I've seen exactly ZERO made by AI."

      Toupee Fallacy.

    • Re: (Score:2)

      by Hentes ( 2461350 )

      I mean, the Play Store has always been flooded with shovelware, AI may actually raise the average quality.

    • Because without recurring revenue you're not going to be able to make enough money to keep your software business afloat. I watched several software companies in the 90s and early 2000s go tits up because they just sold their software and everybody bought it and it worked and it was fine and it was great but nobody ever bought more software because the software they had was fine and great so the company eventually sold to everyone who could buy and they went out of business.

      If you're going to run a serv

    • Where is the shovelware? Where's the killer app?
      Shovelware requires some form of userbase. That doesn't exist anymore if every software solution any normal person can think of is just one Google query and one URL away. The Web has won. Nobody will go through the trouble of even installing software these days in most cases.

      • I still feel like fat apps kick the dogshit out of web apps. I fucking hate web apps. They are low quality shitware most of the time. I'm not necessarily even saying your wrong, though. Just expressing my own personal preference, but I probably don't matter: I'm old.
  • Just what we need: the ability for anyone to publish a vibe-coded Android app.

    Brings to mind a famous joke from George Carlin [goodreads.com].

    • Re: (Score:1)

      by ichthus ( 72442 )

      Just what we need: the ability for anyone to publish a vibe-coded Android app.

      Publish? Probably not -- at least, not YET. Develop for personal use? Absolutely. Why not?

      • Then, they share it on their Facebook and Instagram and whatever other places, and it goes from the app leaking one person's data to it leaking 50 people's worth of data! How wonderful!
        And, this is what happens when you take chunks of other people's code without understanding what the code actually does and paste them together and change what Visual Studio flags as wrong, and compile it and jump around all excited, "I built my first program!"

        Develop for personal use? I'll believe that until I see that the

    • Vibecoding and other AI slop will end up creating a mess of unmaintainable software. However, the summary says "the resulting creations, for now, are only meant to be used personally". If the use case is personal apps only, I think it's pretty great. I would never want to download a random vibecoded app from the store though.

      • If the AI can code the stuff you prompt it for: it creates the same code a human had, based on the code the AI is trained with. So: the maintenance is the same.

        And: you do not maintain AI created code manually. You tell it what is wrong and let the AI fix it.

        Are you stupid or what?

        No one who has any clue is going to let an AI produce 30kLOC code, and then fixes the problems by hand. That would be utterly idiotic.

        • Maybe I'm not enough of an AI believer. Maybe we're talking different use cases.

          For enterprise software, you may have senior developers in 10 years that can guide AI to fix issues. I still think that the person maintaining it needs to understand the code, in case AI cannot fix it correctly (that's where I'm not enough of an AI believer).

          On use case, this Android App thing is not enterprise software. A lot of the vibe coded software we'll see will be designed by a random person, who may or may not let th

    • > Just what we need: the ability for anyone to publish a vibe-coded Android app.

      No - AndroidStudio only lets you built the apps. To publish, you need to get through the labyrinth that is the Play Store. There you need to jump through several hoops to even get an account, and then you need to jump through several more hoops to publish an app to the test store. Getting to production... yes, more hoops. Oh, and then you *must* update something every 6 months or they close your account, leaving you with noth

  • Does this produce signed binaries or do they require sideloading?

  • A few months ago to use Copilot to create an iOS app in Swift (a language with which I have zero knowledge or experience). And it got a few things working, however as the app got more complicated it kept failing to compile a working app.I then went to claude and gemini and both faced similar challenges as things got complicated. Is this still true as of today? Do these tools actually let someone with no experience make a more advanced app?
    • It very much depends on what is classified as advanced there. It's more than capable of throwing together an obvious AI front end. It's reasonably good for adding the right permissions. It still completely falls apart for anything remotely out of the normal click button call api load new page and show data. For example I was mucking around with esp32 and android BLE. Something that is very well documented on all three sides, certainly not an advanced feature yet it took hours of debugging something that I c

      • It very much depends on what is classified as advanced there. It's more than capable of throwing together an obvious AI front end. It's reasonably good for adding the right permissions. It still completely falls apart for anything remotely out of the normal click button call api load new page and show data. For example I was mucking around with esp32 and android BLE. Something that is very well documented on all three sides, certainly not an advanced feature yet it took hours of debugging something that I can manually write in under 5 minutes. I will give it credit it did suggest using a feature I hadn't even heard of that got around a problem. However that problem didn't exist when coded by hand. Despite software developers raving over it online I'm still yet to find it that capable outside of a search engine. It's hundreds of times better at UI than I am but that's a looow bar!

        Thanks for your reply. That's kinda what I was experiencing. Basic stuff worked, AI write code to access HealthKit and load into the app some medical data. However, once it got to parsing the data for use it was a complete failure inclusive of a doom loop of tries. Exhausting.

    • Re: (Score:2)

      by allo ( 1728082 )

      If you know how to program, yes. If you do not, you will reach a certain level and after that it will stagnate as you fight with how to extend it, get the LLM to remember all decisions, or notice that the base architecture wasn't built for what you want know. I mean most laypeople have no idea about what's refactoring, how to build extensible architecture, or when to decide to rewrite when the current architecture is no longer suited for what you want now. A LLM is like a genie, it gives you what you asked

  • They all look and feel like they did that for many, many years.

  • Worth a look (Score:5, Informative)

    by ZipNada ( 10152669 ) on Wednesday May 20, 2026 @12:48PM (#66152823)

    This web-based development utility looks interesting and I tried it just now with this prompt;
    "I want an Android app that can scan for local WiFi access points when a button is pushed and list them."

    It gave me a nice tab completion, "It should also display the signal strength for each access point" which I accepted. And yow, it built the complete app in under 10 minutes. It's running in an emulator in the browser and showing mock access points, visually looks very good. I am officially giving everyone permission to use my amazingly awesome prompt, try it for yourself.

    When I ask how to install the app on my phone I get; "To run this application as a fully functional, bare-metal hardware app on your physical Android phone, you will need to build the provided native Kotlin source code using Android Studio" and there are some pretty complicated instructions. That will probably be a roadblock for most people and it would be a PITA for me too.

    However I'm seeing that Gemini is now integrated directly into Studio which is extremely handy. I can probably use the same prompt there and get a complete installable app in one step.
    https://developer.android.com/... [android.com]

    • Re: (Score:2)

      by Njovich ( 553857 )

      However I'm seeing that Gemini is now integrated directly into Studio which is extremely handy. I can probably use the same prompt there and get a complete installable app in one step.

      Honestly that integrated gemini in android studio is not that good. But if you point antigravity to that directory it would implement what you described just fine. Basically all of these AI coding tools could do this for well over a year.

    • Re: (Score:1)

      by Turlian ( 859450 )
      "To run this application as a fully functional, bare-metal hardware app on your physical Android phone, you will need to build the provided native Kotlin source code using Android Studio" and there are some pretty complicated instructions. That will probably be a roadblock for most people and it would be a PITA for me too."


      You have to click on the "build an android app" button. Then it'll let you push the APK directly to your phone. I struggled with that for like an hour, because it's happy to create

Slashdot Top Deals

We are each entitled to our own opinion, but no one is entitled to his own facts. -- Patrick Moynihan

Close