If you are a software engineer, yes.
Otherwise, no.
However.
It depends what you mean by app
If you mean something small, a page that does one thing, a form that emails you, a little calculator, a tool only you will use, then yes. Describe it to an AI and you will have it working super fast, you won't even believe it.
The other kind is an app that is actually worth people using. An app that handles accounts, payments, roles, notifications, reports, an admin side, and probably multiple integration endpoints.
What you will get
If you try to build something like that while fully relying on AI without having the technical knowledge, what you will get is the happy path, the main use case, the one you had in your head and prompted it to the AI, and it will look good, generic, but good.
Then you click another button, and it doesn't work, or you try a weird input, or you add a second complex feature and it breaks the first one. You ask the AI to fix it, and it fixes that and breaks something else, because it doesn't know what the finished product is supposed to be. It only knows what you said, and nobody can say all of it up front.
You can get about 80% of the way there on your own. The problem is the missing 20%: the edge cases, the errors, the security, the deployment, and the connections to all of those other endpoints. And what app can function without those?
AI will build what you describe. The gap is everything you didn't know to describe.
Why the background matters
That's why it's important to have the background knowledge. Knowing what to test for, how to deploy, how to integrate with other systems, how to structure the data, and what happens when a payment fails halfway through. AI can help with every one of those once you know to ask, but for the most part, it won't tell you what you should've asked.
Without that background, you end up with an app that mostly works, plus a list of issues you can't explain, growing faster than you can fix them, until you are overwhelmed and give up. I have watched this happen, and it's a painful way to lose a few months.
What to actually do with it
Use AI. A lot.
Ask it to explain things, ask it to teach you. I have been using AI to build me artifacts, small interactive pages that explain one thing, on topics I want to know that I otherwise would've found hard to learn.
Build the happy path yourself. It's the best way to figure out what you actually want before you pay anyone to build it, and a working demo is worth more than any document you could write. Show it to your customers. Show it to whoever ends up building the real thing.
Then, when the demo needs to become the real thing, you have two options.
Learn the job. If you want to do the whole thing yourself, you can, and AI makes learning it faster than it has ever been. It just doesn't make it optional.
Or bring it to people who already did. That is what a studio like ours exists for. Years of knowing what breaks, what to test, how to deploy, and how to plug into the systems you already use, so you don't have to spend those years yourself. Show us what you built, or just the idea, and we will tell you what it takes to finish it.