Summary
- 73% of Flathub apps labeled as “AI Slop” were abandoned within months.
- Many were deleted or received no updates after release.
- Analysis suggests the Flathub app AI ban helped spare reviewers and curb quick abandonware.
Probably people who thought they could make an app with AI, realised it’s not possible to make good software without actually learning how software is made, and never opened Cursor or whatever again.
It’s easy to slop together the 10 Billionth remake of Pacman, Minesweeper, Kanban board, ToDo list, etc.
Actually making something innovative, unique, interesting, and useful is tough. It requires skill, knowledge, creativity, and passion to keep at it long after the high of vibe coding your first solitaire game fades away.
If LLMs are blockchain 2.0, then Vibe coding is its NFT fad. Just a way to take advantage of useful idiots who think being a software engineer consists of typing code all day as fast as possible.
I the clearest indication of this is that we have not seen any products to displace existing market leaders.
Of course they are. That slop is unmaintainable.
Behold, the future of software development: disposable slop!
Why maintain software for decades when you can just abandon it and make a completely new one 3 months later? Don’t need to maintain the code if you just throw it away and build a new one instead.
And all it will cost is a massive amount of AI tokens and environmental damage.
At least it doesn’t come wrapped in plastic, though I guess big tech will figure out a way to do that too in the next few years
Hopefully we will see a lot less of it with the cost of AI going up. Eventually it will just be the big companies that can afford it.
That can afford it for now. Shit is gonna make their whole ecosystem a fragile jenga tower in no time, even the biggest ones.
cost of AI going up
The highly subsidized Anthropic and OpenAI subscriptions will go away, but open weight models are getting much better over time - GLM-5.2 and Kimi K3 are both very good.
The most expensive part by far is training the model. That’s why OpenAI and Anthropic are losing so much money (well, that and the subscriptions).
With an open weight model, someone has already trained it, and you just have to cover the cost of inference, making it a lot cheaper. Any company (or individual!) with powerful enough equipment can host the model, which means there’s competition in terms of price, compared to something like Claude Opus where the only four hosts (Anthropic, Amazon Bedrock, Google Vertex AI, and Azure AI Foundry) use Anthropic’s pricing.
I agree that AI slop is bad, but I don’t really agree with their method of determining if an app is abandoned (no updates in 2-3 months).
Sometimes apps don’t need updates. Some Linux apps don’t receive major updates for years. I used backupninja (https://0xacab.org/liberate/backupninja) for long time even though sometimes there’s 2-3 years between releases, because it did what I needed. (I switched to Borgmatic at some point though).
I didn’t update one of my apps for over a year because it already did everything I needed, and I wasn’t receiving any major feature requests. There’s still some bugs I need to fix but I just haven’t gotten around to it yet.
I update my whole complete Linux box like twice a year. Thrice when I’m bored.
These were new applications and so we can assume over time they need updates. Or at least it shows how much interested developers are working on it. It would also be nice to have a random sample size of similar applications without the Ai to have a comparison how often they get updated.
