Dwarf Fortress' creator says the industry's in shambles over AI

I'd argue that supply isn't really increasing. I mean, are we seeing 10x as many movies being released? NYT is now 10x as thick? Every source of quality has upped its quantity by 10x?

Not at all. Seems output in this brave new AI era is more or less the exact same as it was in the dark ages. The only output that has gone up, logarithmically at that, is spam and other low effort fodder.

I also want to push back on the idea that content is "solved." Consider the fact that you could be born in 1810 or 1910 or 2010 and in each of those eras, have more content available for you to read than you could ever read in your lifetime. People tend to enjoy contemporary stuff vs exclusively reading classics, and therefore there is constant demand for new stuff of quality in principle.

I think writers who are trying to become truly gifted writers aren't going to be out of demand in the future. I mean after all, technology has come up with dozens of new ways to commodity just about every medium of art there is for mass production, and yet, for every medium of art there is there are still sought after and celebrated artists.

Provenance has value. That is something that technologists often forget coming from engineering backgrounds, where everything they interact with is commodified and provenance doesn't matter at all, only the binary of if the spec is met.

This seems to extend even beyond the arts. I would say that the race to add more features to web and mobile apps will face a similar obstacle.

How long can we continue making 10x more features just because we can, until somebody realizes that users do not have the attention to spend on them? There might be a shift, but currently, we are too focused on pleasing shareholders.

>people are also demonstrably not that discriminating at what they spend their attention on now matter how unsatisfying the experience was, once it's spent it's gone

I've refunded more games than I've kept on Steam over the past decade. While there's now a much wider pool of games, there's not necessarily a much deeper pool of games. Since trash sells and since marketing a game has been absurdly difficult recently, I can imagine that there's plenty of people who would have made a good game, but who instead (quite rationally) decided against it.

On the flip side, the firehose of trash has pushed me to mess around and make little demos, so who knows?

Let it collapse get rid of patents and copyright. They are holding us back.

People will assume that means that you can copy and distribute for free and profit without work. You already can.

The business model will have to shift to collecting money BEFORE making something rather than trying to collect it AFTER making. You want a new Christoper Nolan film on his new idea, invest in it. That investment is a risk. Once made the money is made people can copy and distribute however they like. In a cinema, on a DVD, on a streaming site etc.

You've pointed it out yourself, we have these old notions of economics that we apply to a new world where it simply doesn't make sense.

As long as AI is only able to create crap that people are embarrassed to admit they watch (which I think is likely IMO just given how it works) I think we'll be mostly fine. Trash TV didn't ruin the television industry and mobile shovelware didn't ruin the gaming industry.

> a unit of digital media is not a commodity good.

Indeed, so there is no limit on how much of supply can simply be ignored, so 100x every year doesn't need to push anything off the rails.

And in reality there is no 0 cost, the marginal cost isn't about pushing bits in the wires

> something will have to change if we want to still have a functioning world where people make good art and others are enriched by it.

Unfortunately this is completely opposed to the goals of the companies creating AI. An org like Google for example couldn't care less if people are viewing quality art and having their lives enriched by it, but they love it when people spend endless hours viewing slop on youtube.

If we as a society want more of the former it won't happen with the help of leading AI companies, it will only in direct conflict with them since the two goals are in direct opposition.

"[I]n an information-rich world, the wealth of information means a dearth of something else: a scarcity of whatever it is that information consumes. What information consumes is rather obvious: it consumes the attention of its recipients. Hence a wealth of information creates a poverty of attention and a need to allocate that attention efficiently among the overabundance of information sources that might consume it."

Herbert A. Simon

Simon is a fantastic author that everyone alive today would to well to read. He gets misrepresented by Karp in his manifesto:

"He envisioned that by the 1980s humans would be essentially relegated to kinetic tasks, confined for the most part to labor that required movement in the physical world."

Hm, doesn't sound like Simon. Let's see Karp's notes on that source:

"He envisioned that: See Simon, New Science of Management Decision, 38 (arguing that “men will retain their greatest comparative advantage in jobs that require flexible manipulation of those parts of the environment that are relatively rough—some forms of manual work, control of some kinds of machinery”)."

Ok. Seems to support Karp's interpretation. But let's follow Karp's recommendation than and see the source:

"I have done some armchair analysis of what this proposition means for predictions of the occupational profile twenty years hence. I do not have time to report this analysis in detail here, but I can summarize my conclusions briefly. Technologically, as I have argued earlier, machines will be capable, within twenty years, of doing any work that a man can do. Economically, men will retain their greatest comparative advantage in jobs that require flexible manipulation of those parts of the environment that are relatively rough—some forms of manual work, control of some kinds of machinery (e.g., operating earth-moving equipment), some kinds of non programmed problem solving, and some kinds of service activities where face-to-face human interaction is of the essence. Man will be somewhat less involved in performing the day-to-day work of the organization, and somewhat more involved in maintaining the system that performs the work."

A much more nuanced and qualified take. Published in 1977, I would say that in may ways work and organizations in developed economies largely evolved like he foresaw given how the service sector came to dominate the economy and so much human effort now goes into software development, configuration and supervision instead of pushing paper with their hands into and from drawers as it was done in the 1970s.

metal gear solid 2 predicted all of this 25 years ago

Supply has just become near infinite, that means all of the value is now in curation rather than creation, and to pick out the good stuff is now more work than it ever was. Good luck finding the next Jimy Hendrix, Picasso or Doris Lessing.

The article starts by video companies being as profitable as ever. Issue is not with economics not being there.

"art"

If you take away the cynical edge, though, if this actually happens we're talking about an age of mom-and-pop businesses in industries that have not traditionally had them. It can't/won't be all software, but mom and pop stores were only about serving their neighborhood to begin with.

Previously on “companies just aren’t run that well anymore”, HN considered this analysis of Boeing’s collapse following a reverse-takeover by an acquired, collapsing, competitor that ended up kicking out the quality-focused management in favor of profits: (6 years ago, 281 comments) https://news.ycombinator.com/item?id=21971040

I assume a similar analysis exists somewhere for the big names in gaming that financial engineers have similarly destroyed other formerly-great companies; ZA/UM comes to mind as one example.

> Does it seem like companies just aren’t that well run anymore?

If you mean the games industry, it was always poorly run. I worked in games in the 90s and 00s it was a cottage industry of nerds, many of whom had success with an 8bit game in the 80s (where they did everything: a one person team), to then have to run large teams of devs/artists/designers/etc without any leadership experience and more complex hardware.

Happy days :D

Companies haven't been well run in decades, if ever, especially American ones. I've worked for about a dozen companies in my 30+ year career. None have been well run, most are out of business. I do design services now, and have worked with dozens of companies through that, and the pattern continues.

We don't have leaders, or people that can make hard decisions, or think for themselves. They all follow the same book of the day business trends. If you do think outside the box you aren't a team player and are rocking the boat.

I've worked for a few companies from 2000-2026 (2 startups, 2 medium companies), and overall, I would say they all had decent to great leadership. None of them were out to only maximize next quarter, all stressed being 100% honest with the customer, even if it meant disappointing them (i.e. we can't quite make that target, here's what we can do - this was for a consulting company), etc., and they were all trying to make the company grow AND be sustainable.

One startup didn't go too far (folded after 3 years), one grew to ~20 people, lasted for maybe 20 years total, I left after 2 years during the dot com crash.

Both of the medium companies are bigger now (I'm still working for one), and have generally positive outlook. I mean, you can't consistently be #1 in your industry when there are 2 companies with combined > 50 or 60% market share and roughly equal - some quarters, your competitor's 2-3 year plan will finish and deliver better for that quarter than your previous (now "old") product that is in the middle of a 2-3 plan, etc..

So they are out there, but it's RARELY in the "high profile" (FAANG/whatever) "move fast, break things" companies that you hear about a lot or that everyone wants to work for.

A friend said <ride sharing company> was like Game of Thrones come to life in a corporate setting, for example. Yet, some people got rich off it, I'm sure.

I was working at a web shop in San Francisco when the Web 1.0 bubble burst. The period leading up to that was characterized by company execs buying themselves all kinds of expensive new cars and toys, and a kind of hysteria on the part of clients that their brick and mortar businesses were being left behind if they didn't spend gobs of money on websites that had all kinds of gizmos and information but were, essentially, just brochures with a contact form. The FOMO of "AI", whatever that means to individual decisionmakers, feels similar to that time. As does the grift. Every plumbing company and real estate agency back then wanted to say their business was online. Not being online was being left behind. Now they all want to say they're AI powered. Well, business owners and managers need to sound like they know what they're talking about, and brag about whatever they think they've gotten for their spending, especially when they don't have a clue whether it pencils out or not. That's why they were happier back then with a website they could say they paid $30k for than if they'd somehow gotten it done for a tenth of that.

Were they ever well run?

I'll give you my diagnosis, which is derived from this: https://www.semafor.com/article/04/27/2025/the-group-chats-t...

The thing to realize about the SV executive class is that they have constructed a closed ecosystem around them where they only talk amongst themselves and a small number of supporting actors who say what the execs need to hear. A CEO will talk to other CEOs he will not talk to his skip reports. Alternative viewpoints or dissent are not part of this social circle, and despite their 'conterarian' self-image, there's an immense amount of social pressure to conform.

You can can see hints of this outside - for example this site used to see YC folks as regular commenters, but no more. Twitter was changed so that only views favoured by Musk show up prominently in replies. These people certainly do not encounter normal people in their day-to-day lives.

In this information environment you see extremely rapid horizontal spread of weird, stupid and esoteric ideas - one CEO does a layoff, all of them will do layoffs. One turns MAGA, they all turn MAGA, etc. One guy starts yapping about some snake oil, they all yap about snake oil. There no negative feedback loop there.

Look at Strange Scaffold for a game dev company built around being a sustainable business where there aren’t any single ambitious “Bet the business” releases.

It seems like Google, Microsoft, and Apple have all switched into "extract max profit" mode. They don't care about quality or even features anymore.

And it comes from the top: the CEOs are not caring anymore, so their lieutenants are not interested in doing any work either. So the companies get turned into fiefdoms without any central guiding direction.

Companies are on purpose poorly run during the "gold rush" phase, which is where AI is right now. Everyone wants to not be left behind and velocity is placed ahead of profitability and especially ahead of quality.

What happens next (eventually...) is the bubble popping, which for AI will mostly look like established companies pulling back on AI hiring and especially cutting AI SaaS solutions unless they are clearly providing value and improving the bottom line.

What we think of as quality will come in the next phase, once the infrastructure woes get sorted out. Web 1.0 was pre-cloud and websites crashed all the time, no one could handle traffic spikes, etc. After a crash, stability becomes a market advantage, i.e. why Facebook won and Friendster died.

In the 1990s, the adage in game development was that you knew you were a game developer when you had sat at the same desk, had seven different positions with three different companies over 30 months, brought in all your own equipment and still hadn’t gotten paid.

So, the game development industry really hasn’t changed that much. But the way we talk about things certainly has changed.

Generation X had a lot of bad things written about us, but we could write and we could think. We were also significantly tougher than the modern generation.

Things weren’t easy and running companies had always been hard. We didn’t think in terms of ‘employees just collapse’ because we were tough enough that that would have been silly. A few may collapse and that sucked for them, but for the most part we did our shit and lived our lives. We didn’t expect lives of leisure or for companies to speak to our personal missions. We had jobs and we had lives.

There were massive layoffs back in the day too. They sucked. But employees didn’t collapse, the world kept spinning and honestly, having been through those years it’s very very hard to relate to the young now. Younger generations do not have a monopoly on struggling or on shitty bosses, but you sure talk about them too much.

I don't go back quite to the 90s, but this has been true across my entire experience working in a few big multinationals (including one in tech).

Companies will do what Wall Street wants (Wall Street likes layoffs), and at a certain level, they don't want unemployment to get too low, lest their workers gain too much leverage.

> Does it seem like companies just aren’t that well run anymore?

They've never been well-ran, but the laws of physics and economics put hard limits on the amount of stupid bullshit that they could produce.

Now that those limits have been largely removed, an exec at the head of a company in the post-AI age is a chimp with a machinegun.

Anything they ask for can and will be built.

Companies under a capitalist mode of production are funddamentally incompatible with creative output because companies want a reproducible formula that they can scale. This is why we get endless sequels, annual FIFA, CoD, etc releases, adaptations of successful IP and so on. Creative output, be it music, movies, TV, games or books, doesn't scale.

There is an insatiable appetite for increasing profits. Eventually that means raising prices and/or cutting costs and the most significant cost is labor.

Game studios can only sruvive and keep excelling as long as they can maintain their culture and key talent and avoid getting finance people and accountants in charge. This is one reason why acquisitions usually mark the beginning of the end.

We're coming up on the much-ancitipcated release of GTA6. People might say that Rockstar has thus far avoided this rot but I kinda disagree. What made GTA magic for me was the social commentary and satire. I personally thought GTA5 lost its soul. Every previous GTA game I'd replay just to have fun, even if it's to do dumb stuff like causing mass pileups and explosions on the freeway in GTA:SA. I can honestly say that I never once played GTA5 again after finishing the main story and I also almost quit before I did. I have zero interest in the online crap but it's a huge cash cow for Rockstar.

So I imagine GTA6 is going to be visually stunning and I'm sure many will enjoy the new era of online gameply but I am fearful it will be even more soulless.

So, back to your point, I think the real problem is that the game devs need to own the studio. That's the only sustainable model. One could even call that the workers owning the means of production.

I think they just want, broadly, to divest from the American tech work force, whether that's via AI or outsourcing. If "the remaining employees just collapse" it's a data point in support of the thesis that developing software in HCOL markets is unsustainable, and if they don't, margins are better!

So I think decision-makers are seeing layoffs as win/win.

Netflix disruption might be an apt metaphor, since it did severely damage the industry and reduce the quality of its output. Those Hollywood execs were right, even if they were ultimately doomed. It's another step in the Sheinification of everything. If it happens to software too, it won't be a victory for anyone except the owners of software companies.

Who's we? You're certainly not speaking for me.

Making things previously impossible possible? That's the business I joined.

Destroying people's livelihoods without care for the consequences while pursuing monopolies, capturing the value as rent seekers, and ultimately enshittifying entire industries?

Yeah, no, I know I didn't sign up for that, thanks.

And if you're thinking "wait that's not AI". Yes. That's precisely what these companies are trying to do.

It's a huge mistake to think that indie teams would make better games if only they could balloon their projects to AAA-size.

Some of the most meaningful and engrossing gaming experiences of my life have come from 1-3 person teams, especially over the past few years. Meanwhile, the vast majority of AAA titles just feel like time-filling entertainment chow to me: fun-shaped but ultimately pointless. I would strongly argue that artistry and auterism is nearly impossible to preserve beyond a certain game size; the experience can't be focused by definition.

In any case, many of the most talented indie game devs are rejecting this technology outright.

I'm old enough to remember when AI meant a certain things, and games hijacked that to mean "some routines that make NPCs do things" and it's just gone from there. Gamers at the time were vocal about the "AI" not actually being AI, but they lost that battle.

The current use of AI for LLMs is a lot closer to the original intent that the "AI" in games ever was. It wasn't hijacked. It was retaken.

They already bifurcate in most any tech related job posting. You have the "AI/LLM" jobs that are full on llm koolaid and pretty vague about what your role in the job even entails, and then there's the "AI/ML" jobs that are just how they are calling the same old ML jobs these days. The title could be the same between the two, but its usually obvious from the post description what the job actually is.

Unrelated but are there any games using LLMs or ML for the NPCs? I really want enemies that adapt to the player.

Are there sources for this? I'm seeing an 11% increase in revenue from their 2025 Q4 report[0], and still a 1% increase in their 2026 Q1 report[1], the off-season for the videogame industry. Hardware revenue's falling, but it's propped up with their Game Pass and game sales.

[0] https://www.microsoft.com/en-us/investor/earnings/FY-2026-Q1...

[1] https://www.microsoft.com/en-us/Investor/earnings/fy-2025-Q4...

edit: Tarn Adams is also an indie game developer, not a game journalist. Dwarf Fortress is highly influential in the indie game scene.

Revenues overall are rising, but the only real growth right now is in China + East Asia, and in Roblox.

https://www.matthewball.co/all/presentation-the-state-of-vid...

The video game revenue growth is mostly mobile game and gacha game mtx, primarily in the Asian markets.

Instead, we got Universal Paperclips ¯\_(ツ)_/¯

You may wish to check if the screen name JackWeinberg is available.

But will people buy it?

(Spoiler: more than you’re comfortable with, yes they will)

More accurately stated, people's reactions are disruptive. The tech itself sucks and doesn't do any of the things boosters claim it does.

Part of the problem is that some investor somewhere making profits is considered the only real marker for success. It’s an insane perspective that devalues all other (arguably more important) aspects of a work’s success.

Profit in video games as an industry is largely the gambling industry hiding under the another umbrella. I'm not even talking about loot boxes. Just old school gambling games like roulette and slot machines but in a digital format. Once you strip that away and loot boxes, the numbers are great for classic games.

The profitable videogames are mobileslop appstorespam games loaded to the brim with microtransactions, as well as the really good games that everyone wants to play, which are impossible to create more of on a whim.

"the industry" also includes "human resources", previously known as personnel or humans.

Any why always call critics "frustrated" in order to depict them as weak and spiteful? Isn't the Dwarf Fortress guy independent anyway?

Which is exactly the problem. All the folks in charge care about is money. Money has absolutely nothing to do with whether a game is fun to play or not.

An eternal curse on the houses of anyone involved with these AIs.

Video games are profitable as ever, but they take years to produce, and the mass layoffs makes me wonder who’s going to be making the games coming out in 2-5 years. Also it has required a lot of write-offs from closing studios to maintain this profitability, and that’s not a trick you can keep up for long.

Profitability can only matter as long as the ecosystem is sustainable; otherwise once the floor collapses there's no more money to be made. A game can be profitable, but when the entire team that made it is laid off or furloughed before launch there isn't a team anymore to follow up. Eventually those people leave the industry and there isn't enough capital made by your average worker to fund their own gaming studio, relying on investors or the rich to parasitize their work in order to create more games.

This is part of the reason why we see endless games as a service despite 90% of them failing. Teams have to do the bidding of the rich, but the rich don't know what's 'fun'.

A sizable portion of the blue collar economy collapses too if the white collar portion stops having money to spend or suddenly has an influx of time to do some of the stuff the blue collar workers were doing for them like lawns, home repair, painting etc.

Its all tied together. The current system cannot function if unemployment goes double digits.

> Blue collar workers and service industry workers aren't in immediate danger of being replaced by AI, so there is a cushion period in which there is still a large number of people who have jobs.

There will just be a lot more workers competing for those blue collar and service industry jobs in the future which will drive down their wages.

"Blue collar workers and service industry workers aren't in immediate danger of being replaced by AI"

I wouldn't be so sure about this; robotics is advancing rapidly. I do think there'll be a substantial market for "human-made" products and services as luxuries, though, so at least some humans will retain these jobs.

> The problem for the CEO is that every other CEO will be able to do the same thing, so what will be the distinguishing factor?

No, the problem is, no one will need the CEO. They'll press the button on their own.

The issue isn’t every other CEO having access to the “make game” button, but who will buy that game when the customers have access to their own “make game” button?

Not to mention “labor they don’t understand” is only possible with specialization and modernity.

Back when everyone was a farmer, yeah sure everyone understood the job

It's possible to be very talented and a crank.

I don't think there's a way to really say a comment like this "with all possible respect" given the (albeit meager) content of the article.

Either way, I'd be much more willing to bet that if Toady had Opus/Sol/whatever we wouldn't have Dwarf Fortress, we'd just have Rimworld: perhaps a technically better game, but not !FUN! in the same way Dwarf Fortress is fun.

There is no universe in which Dwarf Fortress becomes a "cultural milestone." It's core concept and gameplay would never appeal to the mainstream.

The premise that the only thing holding it back is not using an AI seems far fetched. How much game code have you looked at? If it shipped, chances are it's a goddamn mess. People made memes about Undertale's thousand-line switch statement. But it isn't the code quality that makes a game (unless it breaks, if it works literally no one cares) but the design and intent behind it. Using an AI wouldn't result in anything as unique or as innovative as Dwarf Fortress. It would just get something forgettable and mediocre to market faster.

This seems to be miss the topic. Isn't the topic about AI and how damaging it can be? I am not sure why you want to discredit the messenger; the message appears to be correct. That should be a better focus.

> Had Adams been able to use Opus or Sol when he started with DF, that game would be a broad cultural milestone today, and not just a nerd affectation.

Well, that is an assumption. You have to demonstrate this claim. Right now I do not believe that claim.