Well the real problem is the signal to noise, the noise looking almost just right, and people over excited over the novelty of "I can help now!", even if they really can't.

If the thing runs right, then it would be less of a problem, but it almost runs right.

It's actually a lot tougher to iron out bugs that GenAI can't sort out for itself than I usually see with human code. Trying to fix a bug the vibe coder couldn't get the AI to squash is always a maddening nightmare...

and don’t analyze

see: "decomp.wtf turns it back into C, one function at a time, and checks every function byte for byte against the original"

when the C code creates byte-wise identical output, then there is not much need to analyse. The code is already as good as it gets, or at least there is not much room for improvement.

or do anything with the results.

see: "People have, for example, used AI to put Mario into Elden Ring, port games to platforms they were never designed to work on"

I've seen lots of deeply AI fanatic behavior where they are entirely oblivious to how obnoxious other people are finding their output.

This case seems like they did some reflection and internalized the situation, which is nice to see. Most decent folks step back when they realize they are actively shitting on the work of people they respect. Hence AI slop being relatively more popular among sociopathic folks.

Note there is a reasonable difference between AI use and slop, but the "lazy didn't give a shit" nature of AI slopists means they can produce way more volume than the more quality users.

I will say that a whole upcoming generation is pretty much noping away from entering this industry. Can't blame them, as at least for a large chunk of their early earning years, companies will not be hiring them.

Note this theory of "if AI can clone/decompile, then it is fair game" is a pretty baseless "legal theory". People are stitching together incomplete understandings of cases. It is kind of like all those "sovereign citizen" legal theories where people think they have the magical legal recipe only to get slapped hard when they have to try to reconcile that with a real court.

Open weight models can be run by anyone with the hardware, and many models can run on residential hardware. Rich people sure think they are going to control it, but it's not clear that's going to be the outcome.

Is that just a feeling? Because my feeling is that the foss movement is entering a golden age even when you take into consideration all of the slop coding getting posted

To be fair, the slop users are relentless. Trying to even decide when to say yes to becomes a maddening chore if your community is popular and attract tons of random people excited about the novelty of being able to "contribute".

As in other cases, AI is at its worst when you have to deal with other people's output. So closed source can have a reasonably happy time because everyone that isn't on the same page is within slapping distance. But open communities and projects... Oof...

I think it’s just harder to find projects that aren’t slop drops. I don’t want to sound like these communities don’t exist (currently) but it’s hard not to worry about the future.

But I do also believe in the power of communities to organise around a common goal. If there’s a game you like why not start a project and find others that share your goal… just be prepared that there is a lot of fending off of low quality slop right now, and the expectations of the general public for something to get released quickly is at an all time high.

Contributing to a human led decomp project has been one of the most rewarding and enjoyable things I’ve done

AI decompilation is not free software. You're relying on proprietary models, for one thing. And you're starting from a proprietary base program. Sure, you might get something usable at the end, but "free software" means you don't get sued for it either.

While this is a perfectly valid point you make, all these companies have spent the past couple years arguing that all this is fair game as long as AI does it.

"Rogue agents?" Do you really believe that? Which reminds me, I have a bridge for sale in Brooklyn if so. Act fast I'll even throw in the one next to it for half off.

There are currently no laws that apply to this BS directly and that is precisely the problem that we have all had with AI bros for so long, and it is the problem they are about to run smack into, but not before their own toys cannibalize everything they've built into a zombified approximation of itself with bits of other ideas grafted on the side like some Lovecraftian nightmare program. And it'll be just as maddeningly incomprehensible and their ultimate ruin the moment they behold it for what it truly is.

Again, that doesn't mean regular folks won't be caught up in the cascade effects all this will cause, but when the laws go up to rein the bullshit in - and one way or another they surely will - it isn't going to have to be you and me who figure out how to work without AI. Like I said, these companies are talking themselves blue in the face that IP doesn't apply to AI. And now its coming back to bite them.

This is only going to cascade. And when the real court does get involved, the AI bros' own legal arguments will work against them in so many cases for talking out both sides of their mouth all the damn time.

I'm not playing with AI and have gone out of my way to avoid updating my tech with anything AI these days. Linus has made that an exercise in futility given my current OS, but I am nobody and trust that my standard security measures - which are definitely above the average Mac/windows user - will continue to keep me safe from the average threat I am more than likely to face.

I think your sovereign citizen comparison was apt, but even in the name of corporate interests, I don't see it going against precedent when push actually comes to shove; these stupid bastards wouldn't stand for anybody else's hands on their pie and that's what they've inadvertently been arguing for with how things are developing.

People can be quitting for "bad" reasons, but assuming they all are is unfair and unhelpful. That large groups and projects are shuttering over these changes should be met with more than "they're just overacting". Even if they are, the long term damage is real and should be concerning.

Based on the high schoolers my kids associate with and how they have broadly ditched the idea of software development as a viable career to try for.

Also our internship program dropped like 90% in applicants this year compared to last. Most years we do interviews to pick the ones that get slots, this time we just took them all and many groups just didn't get their interns.

I think to the extent AI works, it strongly steers people away from foss collaboration. Why collaborate if the AI can just generate a bespoke solution? If the AI output doesn't work, then other people aren't going to be excited to try to help you make it work among a sea of other broken code.

Maybe post bubble pop. If AI falls to work as well as hoped, a move back to the former status quo. If it does work out well enough, maybe sofware developer goes from six figure to a boring middle class job that doesn't attract so many comers.

What they have so far argued is that it isn't infringement until an operator asks for output that infringes, and then it is the operators fault. I hate that argument, but so far they sidestep the dicey issue by putting the burden on the operator.

The rogue agent thing is bullshit, but we have shitty criminal enforcement and complicit would-be plaintiffs rather than a lack of law.

They did for example rule against that one company ripping off anothers legal material despite the "but AI" defense. The court didn't care one way or the other about the process, they just cared that the output was obviously infringement. AI or no AI.

Oh sorry, sure yeah I wasn't talking about careers. I was only speaking of volunteers specifically. Claude is cheap right now but when it becomes full price I don't think most people doing video game decomps will want to pay for it.

You’re relying on proprietary models, for one thing

The development of the software requires propietary models, not the software itself. If I use windows with a propietary IDE to develop software and make it open source that software is still free, even if the build and development system requires propietary software.

means you don’t get sued for it either.

Clean room rebuilds are legal under copyright law. If you can prove that you didn't see the source and only made something based off observing it's functionality that is allowed.

Heyyyyy! I am not laz...ahh, he's right. Who am I kidding. I'm lazy af.

Yeah but is there really any value in something like that? Maybe for corporate espionage, I guess, but that 250k is still heavily subsidized; these AI companies are hemorrhaging money. What happens when the capex fountain runs dry; is anyone going to pay a million dollars in tokens to get an 80% accurate decomp of an old pokemon game?

Oh I think it is shit, especially reading that 250k and the result was still 20% off of target in a scenario that should, in theory, be right in GenAI's wheelhouse.

Btw the decomp I was referencing was of the original modern warfare 2 from 2009.

Btw the decomp I was referencing was of the original modern warfare 2 from 2009.

lol, amazing. This is what we get out of these datacenters burning tons of diesel. Everything is just so stupid.

I think the AI luddites are playing into the hands of the wealthy, if they aren't actually secret allies.

The problems with AI boil down to individuals having too much personal influence over humanity, no different from personal ownership of factories, banks, politicians, PMCs, and churches.

Everyone telling me and other weak individuals to not use local AI, are basically trying to strangle a new form of democracy in the cradle.

Easy is answering the question "what is 2 + 2?"

Low effort is using a calculator for the answer.

Yeah, more people need to be taught that generative AI models are literally auto-correct like on their phones. It is the gigantic, multi-trillion-dollar, ultra high tech, ultra high powered version of auto correct, but that's what it is. Would they trust something important that, sight unseen?

For those of us who write code, it can be handy to have an auto correct that can suggest chunks of code rather than just the next word or filling in a partially typed variable name. But you have to treat it like a suggestion, and read the suggestion, and make a conscious decision about whether that is how you want it. Giving up your agency is bad, but that is what a huge portion of the population seems to want to do at every turn. Not having to do anything is a nice-sounding option to have, but it's not good for your brain or body.

But on top of that, as an engineer, you are also losing out on familiarity with the design when you give up that agency. I find that when I let the neural net in my head -- the one made out of actual neurons -- get more exposure to the internal workings of a system, it is better at helping me diagnose problems with our product more quickly and intuitively.

My friend. That chatbot isnt giving you individual freedom. It’s owned by a billionaire that is cooking our planet, and it is being tasked with doing the fun creative jobs faster than you.

The end goal isn’t individual democracy. It’s to make you work 10 hour shifts in the lithium mines to pay for your weekly subscription to Claude v47, which will make soulless content for you to absorb in the short period between your mining job, and the second job you need to pay for food and rent.

I use a local AI open weight model, Qwen Next-Flash. I don't pay for a subscription, and it isn't doing the 'fun creative jobs', it is digging through code and building programs.

It is agnostic to political alignment.

Prog doesn't read theory🤣. Hey bourgeoisie, do you own the means of your Markov machine?

This should be the top comment, cc @velma@sh.itjust.works

Embraced, Extend, Extinguish.

Hey, quit gatekeeping quitting!

The development of the software requires propietary models, not the software itself. If I use windows with a propietary IDE to develop software and make it open source that software is still free, even if the build and development system requires propietary software.

That's part of the point, though. Software developed on Windows in a proprietary IDE could've been theoretically developed on Linux with neovim, but software developed using proprietary AI couldn't have been developed in an open source way; there's no meaningful way to replicate its output for confirmation. You don't know how it was made, and you can't produce its output for yourself. You can't continue development in kind.

Clean room rebuilds are legal under copyright law. If you can prove that you didn't see the source and only made something based off observing it's functionality that is allowed.

Theoretically. We're still in a world where the corporations have control over the courts and could very easily spend enough money to get them to rule in their favor.

Besides, I think it could be very easily argued that an AI decomp isn't a "clean room," since you're directly modifying the built binary.

EDIT: also worth noting, Windows and Visual Studio's developers were paid for their work (the developers of AI's training data were not), and Windows and Visual Studio do not have a major negative environmental impact. Not to mention, learning decompilation is a valuable skill that translates to other disciplines, and abandoning it entirely means that we're losing that skillset as a society.

Can't compare code quality for code that was never written. If someone reverse engineers a game that no one was working on with AI that code is better then nothing.

We know AI produces unmaintainable spaghetti, having LLM-generated versions reduces incentive for human programmers to make their own version.

Doesn't have to, I'm fully for having platforms for only humans and if someone wants to make a platform for reverse engineering without AI that's fine by me. Nothings stopping people from playing chess even though computers are objectively better at it now.

The community mentioned in the article did that, found it was untennable, and decided to shut down over it...

The models will still exist. If anthropic goes bankrupt tomorrow the weights for opus 5 will still exist and be sold off to someone, it's not just going to disappear. Same with open weight models, plenty of people have them downloaded and can torrent them worst case scenario if the companies stop distributing them.

That's nice, but what happens in a coupla years when languages and libraries continue to be updated and your model isn't trained for them? How long can you bodge it together to work with until you have to train something new? Who will handle the cost of that?

In general I think you're underestimating the capabilities of modern AI and overestimating the capabilities of your average developer. At this point properly made AI code is better and more maintainable then 90% of developers code.

Does that still count when you talk about people who have been developing for 5 years? For 10?

Yeah it'll make it so you lose skills, but any technology that helps you will. Cars make people unable to walk as far, gps makes people dependent on their phone for directions, hell even shoes stop callouses and make it so you can't walk without them.

Both of those are common skills that can be built up over a matter of weeks to months at most, as opposed to niche professional skills which take years to build.

The skills needed to properly keep an LLM in line will atrophy as you use them less and less.

Even in programming we offload large parts of functionality to libraries we don't know the inner workings of.

Yeah that's why I've always used as little third party code in my personal projects as possible, I've never been a fan of over using other peoples code.

Stack overflow probably atrophied programmers ability to think through a problem and read manuals, but overall it produces better code because, as any SO thread will show you, people come up with stupid solutions to problems and having a source to say what is the correct solution increases code quality as opposed to a programmer thinking through it to a naive solution.

First off, programmers have had forums to ask questions since the Usenet days. Granted, most of them never got near how big stack overflow was even 10 years ago.

These forums are double edged sword for education, yes you can copy and paste, but the best answers were posted by experienced devs clearly and carefully explaining what and why they did.

Because unlike an LLM, they understand WHY they're doing what they're doing.

When used properly I would argue stack overflow, like any other good documentation, can make you a better programmer, not worse.

Oh no, ai killing hobbies and art and fun oh nooo such a surprise if only anybody had seen this comingggg

Your question is unrelated to the statement I made. Does the "Markov machine" care if I own the means or not? Of all the tech advancements made to serve the billionaires, AI is perhaps the only one non billionaires were able to use in retribution, hence why they are scrambling to pass all these regulations regarding it and creating dooms day scenerios (apart from the PR aspect)

Also it is not a Markov machine really (seems like you are using random technical words to look clever), if you have seen Markov chain text generators you will quickly spot the difference.

That is why I think they are scrambling to pass all these regulations. They might even try to push things towards hardware gated LLMs but that would be impossible to enforce globally.

From what I've seen from the AI rollout in my company, it's mostly because vibecoders are not coders.

If you give a vague prompt you get basically scratchpad pseudocode quality, but if you have a well defined spec, clear coding standards, etc. you get a passable junior level output, sometimes better if it's a particularly well trodden process like OIDC.

It's not gonna break new ground without direct help/manual coding, but for gap-filling and simple bugs it's become quite reliable at this point, just not in the end of amateurs.

only one non billionaires were able to use in retribution

With what RAM & drives‽ Are you scaling the monopsony?

A guillotine is a famous political tool. So are machines. Users are politically responsible for how tools are utilized. And in the case of video games, entertainment. Slop machine do two things, make the user think they are decompiling games, and make the user dependent on not learning reverse engineering. So that's 2 awful political atrophies: two regressive alignments.

I dislike genAI as much as the rest of you, but…

If you can’t beat ‘em, join ‘em.

Feeds fire into the problem, great job decomp.dev. As someone already pointed out, not everyone will spend 56k worth of tokens.

I disagree. The way it does that is inefficient and with no understanding. I recommend you read up on things, including this blogpost : https://jpeisach.github.io/blog/ai/legoisland/decompilation/porting/reverse-engineering/2026/10/16/decomps-clout-consoom.html

It has no notion of what data and code would have looked like. Yes itll work, yes the output will be byte matching but it would be unreadable and an unmaintainable mess.

that article spends more time on stuff like "it's about the journey and not the goal", and a little about the quality of the output code. But yes, where the article talks about the quality of the code, it does contradict my claim that "The code is already as good as it gets". But it also points out, that AI can be very usefull if you do know what you are doing, which i actually assumed what was happening. I didn't think that laymen were decompiling games, which seems to be the case here? I assumed that people did the decompilation who already know about that stuff.

As I said earlier, AI probably could’ve saved countless hours by giving another perspective into crash_screen_draw_square. Some people have found that AI can, in some ways, accelerate decompilation while helping them maintain a code base, and still create a sense of fulfillment.

It is indeed about regular people doing decomps without having a feeling for it ( i didnt know what i was doing with wii ios/starstruck either at start ) or development in general. That and people using other people's work for their moment of fame after 'finishing' it with ai.
There is also the decomp.dev being abused as a leaderboard, which makes no sense.
Its a whole shitshow...

Fellow AC fan here, was the AC6 recomp made with AI? I admittedly didn't look too much into it. Hope you're having fun with AC8!

It was made with the help of AI. Can't say how much work was done by the AI and how much by the person in charge of the project. That being said, it works flawlessly, I wouldn't call it a slop project.

AC8 is a blast! Not my favourite AC game, but definitely in my top3.

To that I'd argue that you vet the people before the code. I'm taking baby steps into godot for my own dream project and, while it pains to admit, when I'm having a hard time figuring something out I use deepseek to lay out a tutorial and cite the documentation so I can fact check it.

I don't know a lot about coding, but I know not to trust a bot and I know enough to figure out what most things mean. Two of us to every skilled coder like trade apprentices, you'd still occasionally have to fix a stupid mistake when your Dingus mounts all the fixtures to the drywall, but you're actually doing a lot less on average and you can instead focus on the really complicated things. In turn, the apprentices get to a point of working on their own and the masters don't need to carry the load. Might even save some people from the cult of vibe coders

Note that if I can't readily tell if a submission has had "AI assistance" or not, then I am not particularly interested in playing detective in the pursuit of purity. So in your case it's likely I wouldn't really care if it looks reasonably curated.

But I can understand having a standing policy to cite when you just don't feel like vetting what is more obviously "slop".

Though if your submission basically advertises the AI assistant in use, I'm not a fan. If it vaguely discloses, ideally with nuance, the nature of AI assistance (without promoting specific solutions), then I'd feel better.

I am a bit disappointed in, for example, an issue I just received that could have just said "need to add an explicit requirement" but instead was a needlessly long writeup with a little robot emoji and "created by claude code" at the end. Just tell me the gist and it saves me time, I don't want an AI generated dissertation on every little thing, that takes me more work than the 'unprofessional' blurb that folks want AI to massage.

what would you say is easier for a group of dissenters to set up: a robotized full assembly line (mass production, a major tool of capitalism) or get enough compute power to put together a reasonable open LLM model that can reverse engineer propriety softwares?

and make the user dependent on not learning reverse engineering

Apart from environmental damage and labor theft, skill atrophy is perhaps the only criticism of AI that has held up in the last 2-3 years. I would like to remind you that 4 years ago AI had already run out of "unique data" to train and 2 years ago AI had "reached its limit". So I would take many criticisms with a pinch of salt because they are usually very emotional not logical. Going back to the issue of atrophy you have to consider two cases:

1- A person who knew nothing about reverse engineering now can reverse engineer. In all honesty in %99 of such cases the person would probably not put in the required time and effort to be an expert reverse engineer if there was no AI, that takes a lot of will power and dedication. They would simply not be doing it. So in a sense this is a form of empowerment (with possibly not great end products). In fact if someone with a basic level of know how uses the AI correctly, they can get MUCH better at reverse engineering in a short amount of time then they could have by searching references online. It is hard to imagine how these two would balance to create a net positive or negative effect though. This applies to virtually any field that AI is good at. I think AI is fantastic at IT and by asking it to resolve something then asking it how it resolved it you can learnt a lot of commands and tools you never knew existed. It is up to you to make good use of it.

2- A person who can already reverse engineer uses AI all the time to do the job and loses the skill. The same said person could have also used AI to learn about things he does not know but if they instead just want to use the AI as fully automated code generators this is again saying more about the person than the AI. Usage of AI presents you with alternative routes some of which will atrophy your skills but this is due to laziness not due to design.

In the end software or reverse engineered stuff you get out of full automated AI code factories still suck. It needs a lot of human intervention to be reasonable and the larger the software is harder it likely is. But again this is something that actually might play against capitalism. In their greed many companies are trying to adopt the automated code factory approach (because then they can fire people and generate more revenue) but this is only sustainable if AI keeps improving and can actually come up with mostly bug-free software sometime in near future. Otherwise loss of skill and unmaintainable tangle of software will likely fuck everything up.

Oh dogs I hate that thousand words writeup to say I don't know. You'd think they're getting paid per word...

Well, since they are using GenAI, someone is litterally getting paid by the word...