spoiler
Context: The decline started way before AI.
this post was submitted on 18 May 2025
593 points (96.4% liked)
196
3684 readers
1308 users here now
Community Rules
You must post before you leave
Be nice. Assume others have good intent (within reason).
Block or ignore posts, comments, and users that irritate you in some way rather than engaging. Report if they are actually breaking community rules.
Use content warnings and/or mark as NSFW when appropriate. Most posts with content warnings likely need to be marked NSFW.
Most 196 posts are memes, shitposts, cute images, or even just recent things that happened, etc. There is no real theme, but try to avoid posts that are very inflammatory, offensive, very low quality, or very "off topic".
Bigotry is not allowed, this includes (but is not limited to): Homophobia, Transphobia, Racism, Sexism, Abelism, Classism, or discrimination based on things like Ethnicity, Nationality, Language, or Religion.
Avoid shilling for corporations, posting advertisements, or promoting exploitation of workers.
Proselytization, support, or defense of authoritarianism is not welcome. This includes but is not limited to: imperialism, nationalism, genocide denial, ethnic or racial supremacy, fascism, Nazism, Marxism-Leninism, Maoism, etc.
Avoid AI generated content.
Avoid misinformation.
Avoid incomprehensible posts.
No threats or personal attacks.
No spam.
Moderator Guidelines
Moderator Guidelines
- Don’t be mean to users. Be gentle or neutral.
- Most moderator actions which have a modlog message should include your username.
- When in doubt about whether or not a user is problematic, send them a DM.
- Don’t waste time debating/arguing with problematic users.
- Assume the best, but don’t tolerate sealioning/just asking questions/concern trolling.
- Ask another mod to take over cases you struggle with, if you get tired, or when things get personal.
- Ask the other mods for advice when things get complicated.
- Share everything you do in the mod matrix, both so several mods aren't unknowingly handling the same issues, but also so you can receive feedback on what you intend to do.
- Don't rush mod actions. If a case doesn't need to be handled right away, consider taking a short break before getting to it. This is to say, cool down and make room for feedback.
- Don’t perform too much moderation in the comments, except if you want a verdict to be public or to ask people to dial a convo down/stop. Single comment warnings are okay.
- Send users concise DMs about verdicts about them, such as bans etc, except in cases where it is clear we don’t want them at all, such as obvious transphobes. No need to notify someone they haven’t been banned of course.
- Explain to a user why their behavior is problematic and how it is distressing others rather than engage with whatever they are saying. Ask them to avoid this in the future and send them packing if they do not comply.
- First warn users, then temp ban them, then finally perma ban them when they break the rules or act inappropriately. Skip steps if necessary.
- Use neutral statements like “this statement can be considered transphobic” rather than “you are being transphobic”.
- No large decisions or actions without community input (polls or meta posts f.ex.).
- Large internal decisions (such as ousting a mod) might require a vote, needing more than 50% of the votes to pass. Also consider asking the community for feedback.
- Remember you are a voluntary moderator. You don’t get paid. Take a break when you need one. Perhaps ask another moderator to step in if necessary.
founded 5 months ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Bit of a tautology, that. Presumably for AI to "replace programmers wholesale" it would need to produce human-quality code. Presumably human-quality code would not degrade anything because it'd be... you know, human-quality.
From what I can tell the degradation you're talking about relates to natural language data. Stuff like physics simulations seems to be working fine to train models for other tasks, and presumably functional code is functional code. I don't know if there is any specific analysis about code, though, I've only seen a couple of studies and then only as amplified by press.
I haven't looked into it specifically because it really seems like a bit of a pointless hypothetical. Either AI can get better from the training data available or it can't and then it is as good as it's going to get unless the training methods themselves improve. At the moment it sure seems that there is a ton of research claiming both paths for growth and growth stalling that are both getting disproven by implementation almost faster than the analysis can be produced.
This argument mostly matters to investors itching to get ahead of a trend where they can fully automate a bunch of jobs and services and I'm more than happy to see them miss that mark and learn what the tech can do the hard way.
To be absolutely clear, AI is not "going to put everything else out of business". Certainly LLMs won't. Not even in programming.
My "credibility on this topic" is of zero interest to me. I am not here to appeal to authority. I know you didn't mean it like that, but man, it's such a social media argument point to make it jumped right at me. For the record, it's not that I haven't heard about problems with training on AI-generated content (and on filtering out that content). It's that I don't need to flaunt my e-dick and will openly admit when I haven't gone deep into an issue. I have not read the papers I've heard of and I have not specifically searched for more of them, so I'll get back to you on that one if and when I do.
Anyway, that aside, you are presenting a bizarre scenario. You're arguing that corporations will be demonstrably worse off by moving all coding to be machine-generated but they will do it anyway. Ad infinitum. Until there are no human coders left. At which point they will somehow keep doing it despite the fact that AI training would have entirely unraveled as a process by then.
Think you may have extrapolated a bit too far on that one? I think you may have extrapolated a bit too far on that one. Corpos can do a lot of dumb shit, but they tend to be very sensitive about stuff that costs them money. And even if that wasn't the case, the insane volume of cheap skilled labor that would generate pretty much guarantees some competing upstart would replace them with the, in your sci-fi scenario, massively superior alternative.
FWIW, no, that's not the same as outsourcing. Outsourcing hasn't "often been a bad idea". Having been on both sides of that conversation, it's "a bad idea" when you have a home base with no incentive to help their outsourced peers and a superiority complex. There's nothing inherently worse about an outsourced worker/developer. The thing that closes the gap on outsourcing cost/performance is, if anything, that over time outsourced workers get good and expect to get paid to match. I am pretty much okay with every part of that loop. Different pet peeve, though, we may want to avoid that rabbit hole.
You are saying a lot of things that sound good to you without much grounding. You claiming this is a "widespread and significant issue" is going to need some backing up, because I may be cautious about not claiming more knowledge than I have, but I know enough to tell you it's not particularly well understood, nobody is in a position to predict the workarounds and it's by no means the only major issue. The social media answer would be to go look it up, but it's the weekend and I refuse to let you give me homework. I have better things to do today.
That's the problem with being cautious about things. Not everybody has to. Not everybody knows they should or when. I don't know if you're dunning kruger incarnate or an expert talking down to me (pretty sure it's not the second, though).
And I'm pretty sure of that because yeah, it is an infinite doomsday slippery slope scenario. That I happen to know well enough to not have to be cautious about not having done all the reading.
I mean, your original scenario is that. You're sort of walking it back here where it's just some effect, not the endgame. And because now you're not saying "if AI actually replaces programmers wholesale" anymore the entire calculation is different. It goes back to my original point: What data will AI use to train? The same data they have now. Because it will NOT in fact replace programmers wholesale and the data is not fungible, so there still will be human-generated code to train on (and whatever the equivalent high enough quality hybrid or machine-generated code is that clears the bar).
AI has a problem with running out of (good) data to train on, but that only tells you there is a hard limit to the current processes, which we already knew. Whether current AI is as good as it's going to get or there is a new major breaktrough in training or model design left to be discovered is anybody's guess.
If there is one, then the counter gets reset and we will see how far that can take the technology, I suppose. If there is not, then we know how far we've taken it and we can see how far it's growing and how quickly it's plateauing. There is no reason to believe it will get worse, though.
Will companies leap into it too quickly? They already have. We're talking about a thing that's in the past. But the current iteration of the tech is incapable of removing programmers from the equation. At most it's a more practical reference tool and a way to blast past trivial tasks. There is no doomsday loop to be had unless the landscape shifts signfiicantly, despite what AI shills have been trying to sell people. This is what pisses me off the most about this conversation, the critics are buying into the narrative of the shills aggressively in ways that don't really hold up to scrutiny for either camp.
I appreciate that you didn't mean to say what you said, but words mean things. I can only respond to what you say, not what you meant.
Especially here, where the difference entirely changes whether you're right or not.
Because no, "less human code" doesn't mean "less AI training". It could mean a slowdown in how fast you can expand the training dataset, but again, old code doesn't disappear just because you used it for training before. You don't need a novel training dataset to train. The same data we have plus a little bit of new data is MORE training data, not less.
And less human code is absolutely not the same thing as "new human code will stop being created". That's not even a slip of the tongue, those are entirely different concepts.
There is a big difference between arguing that the pace of improvement will slow down (which is probably true even without any data scarcity) and saying that a lack of new human created code will bring AI training to a halt. That is flat out not a thing.
That this leads to "less developments and advancements in programming in general" is also a wild claim. How many brilliant programmers need to get replaced by AI before that's true? Which fields are generating "developments and advancements in programming"? Are those fields being targeted by AI replacements? More or less than other fields? Does that come from academia or the private sector? Is the pace of development slowing down specifially in that area? Is AI generating "developments and advancements" of its own? Is it doing so faster or slower than human coders? Not at all?
People say a lot of stuff here. Again, on both sides of the aisle. If you know the answers to any of those questions you shouldn't be arguing on the Internet, you should be investing in tech stock. Try to do something positive with the money after, too.
I'd say it's more likely you're just wildly extrapolating from relatively high level observations, though.
Hm. That's rolling the argument back a few steps there. None of the stuff we've talked about in the past few posts has anything to do with the impact of AI-on-AI training.
I mean, you could stretch the idea and argue that there is a filtering problem to be solved or whatever, but that aside everything I'm saying would still be true if AI training exploded any time it's accidentally given a "Hello world" written by a machine.
But that point is not the same as LLMs degrading when trained on its own data.
Again, it may be the same as the problem of "how do you separate AI generated data from human generated data", so a filtering issue.
But it's not the same as the problem of degradation due to self-training. Which I'm fairly sure you're also misrepresenting, but I REALLY don't want to get into that.
But hey, if you don't want to keep talking about this that's your prerogative. I just want to make it very clear that the reasons why that's... just not a thing have nothing to do with training on AI-generated data. Your depiction is a wild extrapolation even if you were right about how poisonous AI-generated data is.