- This was pointed out in a thread: https://hw.leftium.com/#/item/47504047
For those that are curious, it is an alarm app with natures sounds instead of dings and beeps or jingles. As he puts it "none of the other existing apps have good enough sounds or interfaces".
But AI can churn it out in a few minutes, and a human can go in to tweak things manually to their needs.
I recently did it with a little audio editing app. I needed something to edit sound effects for a game in a specific way. Learning about audio programming and whatever would've taken me a lot of time, and I put it off forever since even figuring out where to begin was a hassle. I asked an AI to make a basic app that did what I wanted and I put on a few finishing touches. Took about 10 minutes and works fine for my purposes.
I started programming as a child, although I was more interested in system administration.
Today I work as a high voltage test engineer on new substations. I do partial discharge measurements mostly on green field sites.
I have developed amazing tools for myself, software that I could only dream of having that would have taken years of cat herding and millions of dollars, and still not be quite what I wanted. I am a competent programmer, but I don't have the time. I have the ideas, I even know how to do it, but I already work 50-60 hours a week.
Now I can ping pong my ideas with chatgpt and claude and get something useful out of it, all in the background.
The only thing that leaves a bad taste in my mouth, is that this technology is owned by at best neutral corporations. I want to host it myself, but I can't pay the money, which sucks.
For example, TikTok revolutionized short form videos by introduced a UI where users can explore a lot more videos more quickly, and the cost of showing a bad video is significantly lower (then clicking on a long YouTube video). This eventually led to algorithmically led discovery and content that was created for that format.
You could easily imagine a TikTok for games, where you can instantly scroll between and start playing games with no installation or frictions. Over time games themselves would be designed for that format, hooking you up from the first second (like a TikTok video).
This would obviously change apps and games fundumenltally, just like TikTok did for videos.
Though I agree with you that what you describe will likely happen, I think that is not the future I will enjoy.
Nope. Unrelated company - just looks the same.
Shovelware, but a snow shovel, not a regular one.
Life is fun again. I feel energized and productive with my intellectual wheelchair.
There are a few examples of the types of apps people are making in the comments on this page and it sounds like a lot of the time it’s custom apps for a personal itch that might appeal to very few other people.
Similar to how I have Python scripts on my computer that I won’t publish. They’re not production quality, but I know how to deal with their quirks. I suspect a lot of the apps mentioned in the comments are similar and maybe that’s fine.
I imagine some of the rise is people using AI to clone existing apps too.
An LLM being used to generate the code doesn't say anything inherent about the quality of the app. It just changes the distribution. It's like how knowing a person is a man makes it a lot more like they're into watching team sports, but it doesn't mean they actually are. Or that a woman isn't. As of April 2026, an app's code being primarily LLM written it's a lot more likely to be slop, but there's plenty that aren't slop. Similarly, there were millions of slop apps out there before anyone out there was using LLMs to code.
More and more people will start agreeing with this take and admitting it over time as the Overton Window shifts and it becomes more acceptable. But this has been true for 2 years now. The distribution has shifted, ironically towards a higher % of slop, because 2 years ago harnesses and other tooling weren't good enough yet to allow people with zero coding experience to build an app. It still took more effort, higher barrier. But people were already generating useful code with LLMs, and yes, even creating useful, non-slop, thought out apps whose code was majority LLM-generated.
"Show them! Where are they!?". Look at these comment sections, then take a guess why we aren't linking them everywhere and shouting it from the rooftops. Because it hasn't reached mainstream acceptance, so it's bad for business. People immediately associate it with slop, and they're not necessarily wrong in doing so because as said, 80% is slop. So for us 20%, who didn't vibecode something in a day but worked on it for months to produce something actually valuable, all it does is invite negativity with business impact. In offline, informal settings you'll find way more people who are willing to say it and indeed show.
Increased enablement and higher productivity leads to greater output which will eventually lead to greater need for (some types) of tech workers in the end.
Anybody categorized these apps? I mean, there were literally hundreds of flashlight apps in the Google android store at one point. Are "AI" apps doing variants of one thing, or are they all over the map?
It's never been a problem. Not even where the steep cost is.
A take from Forbes a few weeks ago,
The App Store Is Flooded with AI Slop, Legitimate Developers Are Paying for It
https://www.forbes.com/sites/josipamajic/2026/03/24/the-appl...
Along these lines, an interesting category of work is when I have an LLM do something I could do myself. I totally understand the code, I instruct it all the way, I have it redo things, revise, rejig, etc... But I don't actually write any code. How responsible am I for any of that?
At work there are a ton of small scripts I use for piping data around ad-hoc, and this is often how I do it. Claude can make dumb pipes really well and remarkably quickly with reasonably clear specs given to it. I compose all kinds of specs, reports, plans, etc. using this workflow. And I find myself wondering... How much of this is me? How much credit do I deserve? The code is gone, the outputs remain, and I can't quite tell how responsible I am for the end product. It's a strange experience.
It's fascinating to see so many ideas and so much enthusiasm. I sometimes wonder if the fervor will die down as people realize it's still really hard to make truly fantastic software, but it's hard to say. There's a ton of inertia behind the vibe coding rush.
I also wonder if vibe coding is actually somewhat incompatible with the states of mind and contemplation that's often required to figure out how to solve problems properly. It isn't clear if you can brute force great solutions without putting in the initial domain distillation and idea incubation and so on. I'm sure there are exceptions but I have a feeling it'll never be trivial to come up with truly good and novel ideas for software, and vibing to get there might not make it any easier.
I am old enough to remember old programmers complaining about the wave of new shareware/freeware apps that people made with Visual Basic when that came out. Many of the apps were visually awful because it opened up desktop app development to people with no aesthetic experience.
I don’t see that awful style any more despite those tools for rapid UI creation still existing, did those people get better or did they get bored and move on to other things?
I guess the same will happen with vibe-coders, they’ll get the experience to make better software or their poor quality apps won’t give them what they want and they’ll move on.
There you go. Your apps are here. Even if 95% go nowhere, some will become 1 person unicorns, perhaps 1 - 3 million ARR companies. If you're wondering if anyone gets rich off AI, it's that 5% of people that couldn't write it before - well they can now with a $20 subscription.
See also: dark forest hypothesis and AI.
Another similar but different point: no software development expertise doesn't mean no other expertise. As an extreme example, good luck building tax software with zero tax expertise. This applies to tons of niches.
A third point - lots of this increase in apps is from people who do have software expertise. They're now just able to create things they didn't have the time for, despite their expertise.