In Defense of Giving a Shit
possibilityspace.org
possibilityspace.org
The extent of AI use, which the author does not dispute, sounds really, really mild, i.e. "...they [House House] had used LLMs for "less than 0.1% of the code" to fix errors late in production." And this is still characterized as a "mistake".
I have no objection to this perspective, I am not part of this community and completely respect the efforts of any group to preserve an AI-free space for their particular discipline.
But surely a community built around a standard of 100% AI-free will quickly become quite niche, no? Especially when AI is used in a fashion that is absolutely orthogonal to the actual substance, like in this case where it corrected some bugs late in production.
I do, however, see a massive difference between "AI as tools" and "AI as creative director". I agree with the mindset of "I want to hear stories from humans" and "I want to experience creative expression from humans". But some of the current discussion feels like the arguments we had back in the day of whether Photoshop/digital media/video games were "real art" or not. If a human has creative vision and uses a robot stir-er to mix their paints, I don't think that should invalidate their painting. We can still judge it, though.
Can't blame someone else and take credit for the work when convenient.
You see people brag about how "they" "made" something, with neither the qualifications nor the vision to motivate me to care. If I wanted vibecoded slop, I could just cut out the middle man, like everybody else with a Claude subscription.
But this stuff really does shine when competent people such as myself - an expert, gentleman, scholar, etc. - use it for things like exploring a new topic, rubberducking design ideas, diagnosing problems, and sometimes go the extra mile.
Software projects are full of incidental, boring stuff, and it makes a meaningful difference to be able to say to Claude things like "please compare the performance of these 5 different approaches, and show me your notes". But just like a colleague, it has no clue what your actual intentions are unless you tell it, and you can't communicate those intentions if you don't know them yourself - and crucially, you can't know them if you don't have the expertise to know what you actually want.
This doesn't feel like cheating at all, and I don't relate to the choir of disillusioned programmers who feel demotivated by the introduction of LLMs in their workflow. I mean, maybe their work wasn't all that interesting to begin with? I don't know. I get that it sucks to feel replaceable.
On a different note, AI tools have been great at combating corporate bureaucracy! If the Big Man says that all modules need unit tests with 80% code coverage, you can give the agent that task and then move on to more interesting hard-thinks.
So I'd expand and say that the line should be drawn at knowing how the architecture works, but also we need a pinch of good faith, because it's not very hard to know when you cross the line of overreaching for AI, and it's usually when you hit a point where your prompts start to look like "previous implementation didn't work, try again" or "that's not what I had in mind" or "bug is still there try fixing again"
I've been thinking about this a lot, and I've been going back and forth on it. Because I do feel similarly, but I don't think the premise is itself quite true:
- It neglects that people often don't do things that they want to - and can - do, and that lowering the activation energy can in fact make a positive difference in things happening vs. not. (i.e. "You could've, but you still didn't" happens all the time.)
- It neglects that tokens are, in fact, limited - especially with the best models.
- It neglects that prompting itself is be a skill that can vary across people in a manner that doesn't have a 100% correlation with their ability to understand or do things manually.
- It neglects that time is limited, even the smaller amounts you have to spend with AI assistance.
- It neglects that sometimes the friction is elsewhere (e.g. setting up the harness and thorough testing of the change might have a lot of overhead), not in reviewing the change itself.
- It neglects that the value of a single isn't binary: some person or AI having said "this looks plausible to me, though I'm not an expert" can still be a better signal than nothing, even if a weak one.
The trouble is, of course, that the humans mailing these out often have a very different sense of which way the balances tip than the reviewers do, so what could be true in theory still often isn't in practice...
IMO, this is a wrong way of looking at things if the job is software engineer.
We aren’t the end user who is driving. They don’t need to know how the car works. It takes them to and from work or their kid’s sports event.
The engineer does need to know how the car works. Sure, there are specializations so someone working on one part of the car might not know how another part works. But, they know how these things work together at least at a high level.
I’m noticing that without proper planning and architecture, these AIs go off the rails and introduce horrific complexities that are hard to diagnose and even harder to fix.
For some small app that nobody cares about besides me, I don’t really care about the internals much. I can have the agent churn and do fixes and updates.
For something larger that will have end users who rely on the software to work, I don’t think having the agents burning millions of tokens and many minutes on each task is a good way to work.
The strength of AI is that it can work across many files, reading and writing magnitudes faster than any human. Their weakness is that they need direction, otherwise the result is chaos.
I’ve tried various approaches and honestly, I’m back to spending time up front planning the system and doing a little bit of mixed (manual and AI assisted) coding to set the foundation.
A few days or even a week or two of focused writing concise documentation, with enough foundational code that shows how the system is architected, yields tremendous results down the road. Even the AI benefits from it because it knows exactly where, what, and how to change. Every system is unique, that last 10% that separates it from the generic recipe is what matters.
I am not saying that humans were perfect or that I’m some god-tier engineer. I know my limits, and as an AI enthusiast I have also realized where these tools have consistently fallen short. Unfortunately, we must realize that no system is perfect, and humans need to do boring work like planning and proper guardrails because chaos is easy.
Sorry for the long write up/rant.
Not a question a customer can give a clear answer to.
I don't want to doommax: AI is great, but if we allow ourselves to lose understanding of what our code is doing then we're not doing engineering anymore, we're just doing summoning spells hoping the complexity demon spirit doesn't kill us as a result.
It sounds like this author’s argument is “AI in games sucks for lots of reason but mostly because it robs people of jobs.” I don’t disagree with that in theory but the real question is whether House House would have actually hired another human.
My guess is that they would’ve either delayed it more, or not known about the bugs and released it. They are purposefully small and bringing on even a freelancer introduces a lot of complexity and room for error. So too does AI but ultimately you control the output. You don’t have to worry about it being late, or it having a bad day.
If the end result of using AI is an artful, beautiful game like Big Wall, I’m all for it.
In my opinion the metric shouldn’t be whether AI was used, it should be whether the end result is *good*. I realize that’s subjective but I think we can objectively say Big Walk is a wonderful, artful game. When AI is used and the output is bad - as was the case, say, with those Coca Cola or McDonalds ads - that’s when we should question it.
When it’s being used for artwork or writing - I also think that’s a valid argument against it because it is, as we all know, utilizing artwork and writing without consent.
Is that true for code? Maybe some code but we would all say, I think, it’s much looser around code than creative outputs like artwork and writing.
It sounds like House House used AI like a tool the same way they did whatever engines or coding tools they used.
The metric should be quality.
I am very interested in discussing and reading about the impact of AI on the job market.
However, I think that topic deserves a separate discussion focused only on the broader economic story. When articles start blending every possible anti-AI angle into one generic critique it gets hard to separate the real difficulties from authors who don’t like the fact that the world is changing.
A common story I’m noticing is that established figures with comfortable positions are terrified of AI, while the newer, younger, less experienced entrants are more excited about what it can do for them. It makes sense: If you spent your career grinding your way up a skill and experience ladder you are naturally going to be resistant to anything that threatens your position in the hierarchy of the world. Having hard-earned skills and a history of having done the hard work was a moar for your career and reputation.
Now a lot, but not all, of those old skills is 80% achievable by a kid with a $20 Claude plan. That moat you spent an entire career building became a lot less valuable for maintaining your position, and it happened suddenly over a couple years.
That’s why it’s difficult to interpret articles like this. It’s hard to separate out the valid critiques from the author’s own rationalization for their dislike of AI disruption.
That tracks: younger people lack the experience of seeing ruthless exploitation of employees, before those employees are eventually discarded.
> Now a lot, but not all, of those old skills is 80% achievable by a kid with a $20 Claude plan.
That's a double-edged sword: if the smart kids cannot differentiate themselves from the drooling idiots, they are soon going to resent having done all the hard work for good grades and no recognition or reward.
It used to be that the harder working students got better opportunities. They will soon realise that they have effectively been devalued.
For young people, “their work” is what they can learn to do for a career in the current market.
They’re not like mid and late career folk who are upset that the work environment of the past has irreversibly changed.
They are pragmatic about learning to do what the current job environment needs, not nostalgic for an older job market they didn’t experience.
AI is here and it trivializes what used to be hard. Developing software used to require feats of skill and endurance, and we're used to judging software based on that. Now it doesn't.
We will have to get used to judging software based on the output not on the input. Whether what was made is unique, interesting, creative, not whether it took a lot of time or knowledge.
There is no going back, only forward.
I was trained as an animator at one point. The process was to do the storyboards, layout, key drawings, map out the timings, slug out the dialog, and then send it off to a company in Korea that would in-between and produce the final product.
That Korean studio has been replaced by a machine. Fiscally, that is unpleasant for those animators, but they're really talented and now they can also hire such a machine to implement their ideas if they want to.
Disney was doing this in 1930. In-betweeners and the ink-and-paint girls were the machine. Nasa was doing it in the late 40's. Women were the computers.
Putting software into those roles loses us nothing that is human. As you say, all that matters is the output, not the input.
And if you want to do a project where you think the input is what will impress somebody, because you manually assembled 50k bottlecaps, that's still an available form. Sometimes that works and people care a little more because it was done directly.
All of this sort of already happened, dozens of times. Socks, knives, screws, engines, a million different things that used to be about mastery and are now about understanding.
Understanding is hard. Making that the requirement leaves a bunch of people behind. We're going to have to figure out how to support them, we can't keep having a society that depends on the moral value of work when work is unavailable, but as you say, there is no going back.
We never had a situation where we could outsource unique problem solving at all levels.
You say "Putting software into those roles loses us nothing that is human." but you can have LLM produce design at all levels. I can't think of anything LLM cannot replace humans with, except maybe physical labor, which requires good robotics precision, that we haven't perfected yet.
The main part that's missing however, is that the nature of linguistics (and the core to LLMs' success) is that LLMs are wiring together useful concepts that humans spent thousands of years polishing and relating to one another in speech.
Language has been the code behind all human thought, understanding, emotions, and innovation. It's what enables LLMs to work. LLMs don't _understand_ words, they connect them together usefully, which works because humans mapped useful composable words to useful composable concepts.
And if you think about how human thinking and consciousness is often attributed to language in the first place (not clear if we even would be conscious without language), a lot of human is being delegated.
I'm not sure what to do about it, but I find your kind of explanation incredibly reductive.
That is very interesting and well worded, I use Agents all day so I kind of wish to an extent that they would be almighty, and maybe they will be. But using them so much there is clearly a gap, they actually became worse at writing.
I think, unfortunately to some, LLMS are like perfect operators, great at math and coding, executing on whatever you want, but they do lack some "pseudo creative" ability we might not have a word for. Fable 5.1 genuinely writes poorly. Correctly, but poorly. And as you said, writing codes human thought, well maybe not all of it.
Have you ever read "Stranger in a Strange Land"?
The character of Jubal Harshaw is instructive. He's become very rich because he can just unhook his brain and produce a tear-jerking and profitable story in a few seconds as he lounges around his pool.
He gives his secretaries the gist of the story, they fill it out, they arrange for somebody to make the film.
The point of the character is that what he's doing takes no emotional effort or insight. He's writing Hallmark movies. Everybody who's working for him adds nothing to it.
And we have Hallmark/Lifetime movies already. People were making them for decades before the LLMs came along.
But if you could send an LLM back to 1950 and train it on the pre-existing literature, you couldn't have it generate The Shining, A Scanner Darkly, Piranesi, Perdido Street Station, The April Witch, or Starship Troopers.
-none of the research indicates a mechanism by which you can create a selfhood that gets banged around by time and has feelings about itself, but admittedly, that might change
-even if you could do such a thing, you can't then put the robot in a human body and let it nurse and get potty trained and sent to pre-school and bullied and wonder about its mortality as it starts to think about what its career might be, etc etc etc
We were already making slop, because we already had a big production machine that transforms any input into a "movie" or "a television show", and a lot of the inputs were garbage.
The movie "A Talking Cat?" exists and predates AI by about ten years.
The point is that we were already delegating this stuff to human beings who didn't care about it and weren't bringing their humanity to it. Nothing changes on that side of things except that they don't have jobs to show up to.
So, at least for now, this isn't very different from the previous way of doing things, except that it's cheap enough that some people are spending a few thousand dollars to say "make me a story about a sad orphan" and the machine rips off Charles Dickens and Horatio Alger to make a thing nobody wants to watch without paying a bunch of key grips and best boys and caterers.
This isn't ideological Milton Friedman free markets rahrah. I just can't imagine the alternative in practice: what if mainframe operators decided to opposed destruction, which would prevent mini and personal computers from becoming a reality?
why? you skipped over justifying this statement. especially in an artistic context.
This has really emphasized the difference between developing software and building a business.
Developers everywhere are now doing side projects and developing software that would have taken 10 person-years in 2022. They’re discovering the hard way that software alone does not create a business product that customers might pay for.
Developing software is cheap now, but creating a business remains hard. All of the attempts I’ve seen to use AI to develop and grow a business have only turned real customers away.
This has been the case for far longer than AI has been a consideration. We all know that counting lines of code isn't a mark of quality, and yet we venerate those that put in all the hours to get something done.
I would say you need to go a step farther than that. The output needs to be impactful. Increased MAU (or GH stars, downloads, etc.), lowered support load, lower ops costs, whatever it might be.
These still require skill and endurance to do well.
Vibe-coded projects always are clear as such, no matter how slick they look.
Similar to all other crafts after industrial revolution. This of course will only be viable for a small and niche portion of software market.
> Whether what was made is unique, interesting, creative
Not this.
That is not what technology is for.
If you want unique, interesting and creative - you can have it anytime, just like our ancestors living in a cave or a jungle.
Technology is for something else. You can scratch your head with a hammer - that is not what a hammer is for.
I assume you might associate technology with a more utilitarian perspective, but that's just not descriptive of what humans actually do.
Whatever technology becomes the next revolution (hoping it's not sticks and stones) will be used by humans (hoping they exist) to make art as well.
I think this will always be the front line of competition.
Who has unique knowledge (dare I add taste?!), and is willing/able to give a project the most focus and time?
Would you define everyone involved as "disliking" new technology? Or would you define them as having a "like" for old technology? I think the same question could be considered for communities which continue to enjoy hand programming.
(when I say "value" I'm not talking about making money. Retro hardware generally does not make any money, but it is obviously valuable to those who enjoy it. Someone who hates modern technology might prefer to work with that, but that's not a requirement nor the focus of the community. If instead you have a community which is defined by hating modern technology, what I'm saying is that that is probably more harmful to the people within it than helpful. These kinds of spaces can become _very_ bad for the mental health of those within them.)
Like for example yeah frontier labs are stomping into venerable domains with no respect for their history or culture, and spending billions of dollars crushing worker power. Both projects the tech industry has been deeply invested in for over a decade and I don't remember anything like this backlash to it then so maybe the attitudes aren't the problem exactly?
There are also people conscientiously using LLMs to improve their stances on security and accessibility that would otherwise be outside their budget.
I can think of some other attitudes too. What were you thinking of?
No matter which IGF chooses, I believe awards will ultimately bifurcate: some other contest will emerge for the other side.
I don't think that's bad, nor do I think allowing AI means "not giving a shit."
"Why couldnt they hire a specialist for a few days"
Which is question that I think is willfully ignored by the pro-ai crowd.
AI is useful. It helps people go faster at a lot of things. It's also been trained on a scale of copyright violation that exceeds all modern measures of capacity.
The implications of this are willfully ignored by those who stand to profit from it.
It is simply unethical, unfair, and devoid of empathy for society to simply "march forward" and leave behind the millions of people it spent the last 60 years demanding more of.
These problem needs to be resolved before we can morally and socially embrace AI.
> It is simply unethical, unfair, and devoid of empathy for society to simply "march forward" and leave behind the millions of people it spent the last 60 years demanding more of.
We disagree. This is a luddite point that I spent my whole life disagreeing.
I think it's unethical for society to intentionally block a technology that increases productivity to protect the interests of a smaller class of professionals. I believed that when it was happening to blue-collar workers, and I'll keep believing it when it's me.
And that's how we all wound up with Whispering Earrings. https://croissanthology.com/earring
Maybe you genuinely want one -- I don't think many people do.
It's fine to dislike AI, it's fine to even dislike AI and not be able to articulate a reason, but I'm tired of hackers suddenly becoming the biggest defenders of IP.
Good, hopefully we can ditch copyright entirely and move on forever. I thought that's what everybody wanted?
Plus, the idea that instead of solving the problem with the solve-your-problem button, the idea that the correct answer should be "just throw a bunch of money at it and hope that works" is so silly.
if they don't ban AI code in 3rd party code, you could just circumvent the whole thing by outsourcing a few developers to be part-time 3rd party in name only and let them do the AI development, which would be against the spirit but not against the law.
It's great to have a competition in software writing.
We already have a distinction between people who define and design software and people who implement it. Product owners, UX designers, business analysts, solution architects etc
Perhaps you can expand?
That might overlap with some existing role, for sure.
My main point is that hand written (or even hand reviewed) code will be a curiosity. Like a hand knitted jumper. Only viable in markets as a niche/specialized product.
Also, you vastly underestimate the number of manual steps in consumer goods today.
Then again, I'm not sure it's necessary; so many highly productive programmers seem to dispense with much of that, too. Maybe, in this age of frameworks and libraries, all the patterns are adequately baked in and I needn't worry about refactoring or software design. Hard habit to shake, though.
I don't disagree that AI used to develop artistic assets should be grounds for an entry to be barred from the IGF. I think the code portion, however, requires some specificity.
To me, it's not ok if the code that powers gameplay or artistic rendering of assets is AI generated. That kind of code carries human intent and is part of the art of game development: the dev wants a player to encounter a certain kind of experience.
There's no point or meaning if there's no developer dictating the experience by hand, if you're prompting this you might as well just admit you're trying to make a gimmicky asset flip "object simulator" to make a quick buck.
On the other hand, I think it's ok to use it as a tool and as tooling: trivial tests at all levels should absolutely be automated, figuring out engine quirks or getting second opinions on architecture with AI assistance is fine, figuring out your build and deployment pipelines.
---
Another point that I want to mention: there's a strange form of writing prevalent among both software devs and those in the art space that's stilted; a mishmosh of Tumblr rantposting and slashdot/IRC snark and sarcasm, signaling that they've the mental maturity of a junior in American high school. It's all about emotion (this sucks!) rather than addressing the root of a problem.
This type of person is extremely difficult to deal with as they're the same type of person who encounters a brownfield codebase and immediately spends their time afterward whinging about how "the codebase must be burned and rewritten"; they must be micromanaged lest you set them off.
The author mentions freelancers and such, and how there's a lot of "lip service about how [they're] 'all in this together'" but frankly what I see in the case of House House is a team who wants to keep things within the team.
Regardless of their size, House House is a team on a timeline and budget, why should they be subjected to scrambling budget and conducting an interview process for someone who may or may not be a "brownfield rager"?
The author here however goes further, and claims their argument against AI code is so strong, that they'd lose brain cells by actually making the argument, so we just have to accept that as fact.
They then go on to suggest that using AI for bug fixing is terrible, and hints that people should feel ashamed for enjoying a game that has gone through that process.
I think the issue I see with that stance is that frontier models can now perform asset generation via deterministic code. This is different than the 3D diffusion models of yesteryear. Is that manner of AI asset generation (which is really just through code generation) worse than other AI code in a project (i.e. resolving this bug)? If so, why? (I do not have good answers myself)
There is a little more creativity the further up the architecture and design you go, but the lower level you go, the less creative it is. Indeed for decades we have handed off the mechanical conversion into machine code to deterministic compilers.
Programming is certainly a skilled practice, but it isn't especially creative. Given the same problem, many programmers would produce a similar output.
This is very different to creative work, where the same requirements can lead to vastly different results. The range of art that could be described, as "Draw me the night's sky" is immense.
Generating assets, be that genAI, or procGen, takes away from the creative, and doubly so if you don't have much of a filter on the output and just accept what is given.
Look at procGen: the worlds of No Man's Sky have no soul to them. Compare that to the much lower and less detailed lands of Azeroth, which had so much soul that we're still recreating and enjoying them 20+ years later.
Generative AI can trick us into thinking there is a soul and that something creative has happened when it hasn't. Programming has no such deception.
If you're working with generation but then hand-filtering a million results back down to a hand-picked few, then there is a creative process once again, a curation that is itself an artistic process, so these things aren't absolute, there are always grey areas all over.
I'd be pissed if I entered a contest (and asking for a refund if I paid to enter as this criteria was not disclosed) thinking that I was allowed to make the best game I could make regardless of tech used, only to find out that my max possible score was 66% because shadow judges were biased against me.
It feels like the old reddit moderator meme where the rules are made up and the points don't matter because all it takes is one salty mod.
The fact is, the more I know about a topic the less capable LLMs appear to me.
The less I know about a topic, the more capable LLMs appear.
At the end of the day, save for very specific tasks, there's still a lot of hype.
Ban slop.
Or better yet, don't "ban" anything - judge it like any low-effort crap.
IMO just make categories for AI use and not. The hard part is that people will cheat.
It would be kind of like having a traditional woodworking contest and people show up with stuff made with power tools and cnc routers.