Google has removed the last step that kept most people from having their own app on their own Android phone. If you are 18 or older and in the United States, you can build an app in Google AI Studio and launch it through Google Play with an ordinary Google account. It does not appear in search or in any listing, and nobody but you can open it.
The code was already easy, and the doorway was not
AI Studio has been able to write full Android apps from a plain description since May. What stayed hard was getting through Google Play's door. A new personal developer account has to run a test with at least 12 people for 14 days before an app can go public. Many people who built things with AI gave up on Android apps and made web apps that open in a browser instead.
A web app works, but it does not sit on your home screen like a real app. The new private route needs no developer account. The app still has to follow Google Play's usual policies, and the rollout starts in select markets that include the United States. For iPhone owners and for people elsewhere, a web app is still the way.
What the spreadsheet did to small jobs
The people who got the most from VisiCalc were not programmers. They were people with a small job that repeats and numbers that have to be right: the budget, the quote, the forecast. No software company would build a program for any one of those jobs, because the buyer would be one person. A spreadsheet made that fine, because the person with the job could build the tool.
Most useful phone apps are the same size. A rota for nine staff, a checklist for a delivery route, a roster for a child's sports team: none of them is a business, and all of them are worth having.
For example, a shop owner closes the till each night. She counts cash, card payments and refunds, writes them in a notebook, and types them into a sheet at home. She could describe to AI Studio what she records and what she wants back: a total, and a warning when the cash is off by more than an amount she picks. She could then open the app from Google Play that evening. If it saves five minutes a night, that is about 25 hours a year.
The decision that matters stays with her. She should check the app's totals against her notebook for the first week, because an app written from a description can be wrong in a way that looks right.
The store becomes the last stop, not the first
Google plans to let people share these apps with a small group of trusted friends in the months ahead. If one becomes popular, it will be able to move to the public store after the usual safety review. So an app starts private, spreads to a few people, and goes public only if it earns that.
Most apps will stay at the first stage, the way most spreadsheets never leave the person who made them. For a small business, the second stage may matter most: a tool the manager builds and the five staff who use it can open, with no software project behind it.
The useful question about any chore you repeat on your phone is no longer whether an app exists for it. It is whether describing it would take less time than doing the chore for a week.