Developer questions how much F-Droid contains AI-generated code
A Hacker News post examining signs of LLM-generated submissions in the open-source app repository sparks debate over code quality and developer skill erosion.
What to know
- A developer questioned whether open-source repository F-Droid has been infiltrated by low-effort, AI-generated code, pointing to signs like bypassed code review and large initial commits.
- The post split commenters: some experienced developers defend AI code as superior to human writing, while others worry about the erosion of programming skills and the loss of joy in development work.
- Methodological skeptics challenged whether these signs reliably indicate AI generation versus ordinary human sloppiness, questioning whether the prevalence estimates were too high.
- Experienced engineers noted that while LLMs can produce code faster, they introduce novel maintenance and architecture problems that require skilled human oversight to manage effectively.
The dispute Whether LLM-generated code's superior quality and efficiency justify its adoption by professional developers, or whether the loss of human skill development and coding enjoyment represents an unacceptable social cost. · positions read across 33 posts and comments
LLM-generated code is often better than human code and serves users well; code provenance should not matter if it works.
-
“I've been programming my whole life but basically stopped writing code by hand in 2026. The LLM writes better code than I do, much better.”
edg5000 · Hacker News ↗
AI generation erodes human programming skills, joy in coding, and introduces novel maintenance problems that inexperienced developers cannot solve.
-
“I love writing code to solve intricate problems. But knowing that a faster, and maybe better LLM solution is just a prompt away? Somehow that takes the joy out of it.”
bradley13 · Hacker News ↗
The original analysis overstates the prevalence of AI-generated code; large commits and weak code review are common in human projects too.
-
“70% seems unexpectedly high... Was there maybe some overcounting?”
pona-a · Hacker News ↗
Experienced developers who know how to architect systems and constrain LLM behavior produce better results than amateurs using one-shot AI generation.
-
“People who have previous experience and know how to architect things (and when to stop an LLM from doing something wrong) are doing much better building actually useful things.”
walrus01 · Hacker News ↗
_ZeD_ Student programmer and FOSS app maintainerF-Droid project Open-source mobile app repository
How it unfolded 6 developments, newest first · click a bar or a number to jump articlespostscomments
-
6
Commenters highlight new engineering challenges AI introduces
ModernMech and walrus01 pointed out that while LLMs can write code, they introduce novel problems: building inefficient interpreters, 300k-line scaffolding, and CI systems that fail constantly. walrus01 noted that experienced engineers who know how to constrain LLM behavior are building better products than amateurs attempting one-shot projects.
“Agentic engineering faces all kinds of new problems that couldn't exist before, and need experienced engineers to solve them.”
— ModernMech -
F
https:// tintotint.eu/whacky-corner/f-d roid_slop/ Only apps from one # FDroid cycle were analysed? Was that easy? What will it take to follow each cycle or scan all >4.3K # Android apps? And a poll: How did you find the report conclusions?
2 more of the top 3 · 26 posts in this stretch
-
The author says "Any kind of agentic infrastructure automatically lands an app in this tier as I do not believe it is possible to use AI responsibly from within a coding harness." Which seems like kinda a weird take to me. I can understand a general sentiment of wanting to know how much slop to expect in a project. I can understand a principled…
-
This reminds me of another time... Let's travel back to 1980 and interview a local grey beard:Everything these days is shell scripts! They are even selling them like they have the same value as a C program! csh will never be real C. The kids have no idea what a stack is or how to optimize system calls. The performance is awful and they barely…
-
-
5
Debate emerges over whether code provenance matters to end users
commenters diverged on whether the source of code matters. alienbaby argued "If they work, does it matter?" as long as apps function as advertised and contain no malicious code, while bradley13 raised the emotional cost: "I love writing code to solve intricate problems. But knowing that a faster, and maybe better LLM solution is just a prompt away? Somehow that takes the joy out of it."
“If they work, does it matter? So long as what I paid for does what it says it does (and nothing nefarious..)…”
— alienbaby -
FDroid can be very strange…””” F-Droid is not hosted in just any data center where commodity hardware is managed by some unknown staff. We worked out a special arrangement so that this server is physically held by a long time contributor with a proven track record of securely hosting services. We can control it remotely, we know exactly where it…
1 more of the top 2 · 2 posts in this stretch
-
Same here. I'm still a better software architect that AI, but there is no question that my AI generated and reviewed code has fewer bugs than code I hand write. It takes some humility to acknowledge that your coding prowess is less of a useful skill than it used to be.
-
-
4
Experienced developers defend LLM-generated code quality
Multiple commenters with programming experience argued that AI-generated code is often superior to human-written code. edg5000 stated they "basically stopped writing code by hand in 2026" because LLMs write better code, and 9cb14c1ec0 acknowledged that AI-generated and reviewed code has fewer bugs than their hand-written code.
“I've been programming my whole life but basically stopped writing code by hand in 2026. The LLM writes better code than I do, much better.”
— edg5000 -
Seems like the wrong question to ask. I've been programming my whole life but basically stopped writing code by hand in 2026. The LLM writes better code than I do, much better.
-
-
3
Commenters question the methodology and prevalence claims
Multiple Hacker News commenters challenged the analysis, with pona-a suggesting "70% seems unexpectedly high" and questioning whether large initial commits are actually reliable indicators of AI generation rather than human sloppiness. Others pointed to specific projects like Yubico Authenticator showing no clear signs of AI generation despite Copilot in contributor lists.
“70% seems unexpectedly high... Was there maybe some overcounting?”
— pona-a -
70% seems unexpectedly high... Was there maybe some overcounting?Yubico Authenticator https://github.com/Yubico/yubioath-flutterI actually don't see any significant signs of AI use. There's Copilot listed in the contributor list, but I'm not seeing commits listed under it. Did they wipe it off Github?Some seem to stamp Mostly AI based on weaker…
1 more of the top 2 · 2 posts in this stretch
-
Reminds me of a professor that displayed snippets of Haskell on MS Word in her lectures, formatted by hand. I don't blame her, this was >20 years ago, before Ctrl/Cmd + +/- became commonplace for zoom/font size.
-
-
2
Post gains traction on Mastodon and Hacker News
The article spread across social platforms with nixCraft sharing it on Mastodon and the post reaching the Hacker News front page with 67 points and 72 comments, sparking discussion about the prevalence and implications of AI-generated code in open-source repositories.
-
N
I love F-Droid. I love what F-Droid stands for and I like the freedom that it gives its users. As a FOSS app maintainer I also have nothing but good things to say about the people behind the project. Maybe they’re even a bit too nice, considering how many times repro has failed due to me forgetting to commit before building 😅. But my god is it…
1 more of the top 2 · 2 posts in this stretch
-
> I also noticed a pair of very bizarre apps, both branded with the yellow “Don’t tread on me” flag: DuressKeyboard & UnlicenseLauncher. What’s most curious is that they have been in development for quite some time, yet all the changes are done not with git but through the GitHub web file editor! Someone go find that person and teach them to use…
-
-
1
Developer publishes analysis of AI-generated code in F-Droid
A student programmer published a detailed examination questioning how much of F-Droid contains LLM-generated code, arguing that telltale signs include bypassing code review, large initial commits, and general "low effort all the way down" development patterns. The author noted spotting an app with an obviously AI-generated icon and wondered about the broader prevalence.
“You see, the main allure of LLMs is that they allow the developer to be more lazy. That's kind of the whole point! You just prompt, sit back and relax.”
— _ZeD_ -
first by HN Frontpage, 12d ago
-
What people are saying 17 voices from 2 sites · best of 33 · verbatim
- Sep 16
-
From my understanding of the text, the underlying thought seems to be that slop comes from laziness, and from that perspective I can see it. If you have an "agentic" setup the output will likely be better, but that also makes it more likely you'll accept it as is, without review, and to lower standards than if you would've written it yourself…
-
I agree mostly, but you lose me a bit here: > If the agent is bringing in some quality, rigor, or just action to those things where it wouldn't have existed at all in its absence, I'm not sure I'd object to that. Quality and rigor are not interchangeable with "just bringing some action" which reads as "doing anything is better than nothing." It…
- Sep 15
-
This can be done in F-Droid even. Just have to create an alternative repository that is maintained this way.
-
I've been a bit disappointed in the f-droid app itself... it seems to almost actively hide things you'd want to know before installing something. I've always made a habit to research apps on their website, which has been a bit more forthcoming with details, before installing.
-
F-Droid has always had an app quality problem in general if you ask me. I'm talking broken apps, apps only working on old versions, unmaintained apps, abandon ware, outdated versions up for weeks (that last one is technically due to F-Droid's own build schedule iirc).
-
I guess a different model would to be to try and make a FOSS app store that's more like a Linux distro: app authors don't upload their own apps but instead there's a group of volunteer "app store developers" who do this part. (Don't know if that's been tried before, free idea for anyone who wants to!)
-
I feel this problem grows out of one of the long term limitations of F-Droid: the only metadata the client gives users trying to find apps is how recently the app version was uploaded, and which Free Software "anti-features" the app might contain. In the pre-LLM era this already meant many of the apps which appeared first were people's pet…
-
My first startup job after college I was astonished to find that one of our most productive engineers was a deaf furry who used Windows Notepad as his dev environment. His code always compiled first time and was nearly always right. It really was something to behold."The way that works for you is the right way."
-
how did critics become suddenly concerned with virtue and beauty with AI? Wouldn't it be better to judge these apps on their utility?The authors are sharing their creativity, and spending time & money to publish apps with a free license.I have disdain for how entitled open source consumers are. Before AI, demanding free accounts, support…
-
I have been working on a sub 1500 line rust init system for over a week. Hundreds of prompts. All with a local LLM running on my own GPUs because I expect to build with total sovereignty but also zero dependencies, no libc, no alloc, no std, and a test suite that proves the 20 implemented raw syscalls all use the right values by comparing against…
-
For me the usefulness of a survey like this has nothing to with how effective LLMs are themselves. It's more that when someone's able to produce an app in an afternoon, and submitting the app to F-Droid becomes a checkbox, how confident can you be that they'll continue maintaining the app? Sure if it's open source you can have your own LLM…
-
Every codebase that is being actively worked on (closed/open source) will contain code that's AI generated. With the rising abilities of agents, expectations are sky rocketing in terms of productivity.If you're as productive as an engineer in 2016, you're not at the level that's expected. A 7 day workflow back then should take you maybe a day or…
-
I'm somewhat surprised about PipePipe. I had a look on the commits of the various components and nothing looks out of place to me. Commits look rather reasonable, comments look useful and don't show obvious LLMisms.What are the AI smells there?It would be nice to expand a bit on the reasoning behind the verdicts.
-
With the "intelligence" of code focused and capable llm in the last six months, the main problem I'm seeing now is where some total amateur who has no previous knowledge of coding tries to one shot a project. People who have previous experience and know how to architect things (and when to stop an LLM from doing something wrong that will cause…
-
The AI machine can write better code, it can also write an interpreter which implements function calls by instantiating a new interpreter + entire standard library per function call. Or it will build a 300kloc cathedral of scaffolding and maintain that forever, never writing actual code. Or it will create a CI system that takes 2 hours to run and…
-
It's an emotional problem. I love writing code to solve intricate problems. But knowing that a faster, and maybe better LLM solution is just a prompt away? Somehow that takes the joy out of it. Why spend hours, when you can get an equivalent result in minutes?I will be curious to see how I feel about AdventOfCode this year...
-
If they work, does it matter?Separate from building your own code, ,of course you may have your own standards to apply.But for apps, well, I never had a chance to see how good or bad the code was before AI was about, so why should I care now, so long as what I paid for does what it says it does (and nothing nefarious..)