John Carmack: Calm Down About Video Game AI Tech Demos, Folks
from the dey-took-our-jerbs dept
The fear over industry disruptions due to technological advances is so predictable that we have the entire “buggy whip” analogy pre-built to rebut it. For the uninitiated, the analogy harkens back to when the automobile came into wide circulation and the disruption it had on the makers of horse buggy tools, like the whip for the horses, if they didn’t adapt. While some may have argued that the automobile was a danger to those businesses, the truth is that these new vehicles opened up so many other economic opportunities both for newly created jobs opened up by cars and within the transportation industry for those willing to adapt that it was actually a net benefit to jobs and the economy.
It may feel like artificial intelligence is everywhere these days, but the truth is that we’re still in the very early stages when it comes to how this technology will be used in the future and what effects it will have in all kinds of markets. In the video game industry specifically, we already have examples of how generative AI used in gaming isn’t even close yet to the product of human artists, nor a replacement for the gaming output of human beings. But that isn’t stopping some folks from worrying out loud that use of AI in the gaming industry is a threat to industry jobs.
In some cases, it will be, of course. But that should be a temporary concern, just like the buggy whip manufacturers. More important is what the use of AI in gaming can mean for increased output and as another tool for developers to use to create a better industry ecosystem. Industry legend John Carmack recently made this point when responding to some criticism for Microsoft’s recent demo of an AI-generated facsimile of Quake 2.
Carmack’s comments came after an X user with the handle “Quake Dad” called the new demo “disgusting” and claimed it “spits on the work of every developer everywhere.” The critic expressed concern that such technology would eliminate jobs in an industry already facing layoffs, writing: “A fully generative game cuts out the number of jobs necessary for such a project which in turn makes it harder for devs to get jobs.”
Carmack responded directly to these concerns in a lengthy post. “I think you are misunderstanding what this tech demo actually is,” he wrote, before addressing the broader concern about “AI tooling trivializing the skillsets of programmers, artists, and designers.” Carmack positioned AI as the latest in a long history of technological advancements that have transformed game development.
“My first games involved hand assembling machine code and turning graph paper characters into hex digits. Software progress has made that work as irrelevant as chariot wheel maintenance,” Carmack explained. “Building power tools is central to all the progress in computers.”
It’s the right argument to make, frankly. We’re a long ways off from AI being able to build entire, quality games from start to finish, as Microsoft’s own tech demo showed. Its output was nowhere close to being a one to one recreation of Quake 2. It was much closer to something like allowing players to experience the AI’s impression of the game. In the immediate, this demonstrates that human beings are still very much needed and that the output of AI is more akin to Carmack’s analogy than a job-stealing dev-bot.
But maybe someday it gets way, way better. In fact, that’s probably nearly an inevitability. There’s no reason to think that the long term is one where AI creates video games all on its own sans human handlers that have no input and therefore no jobs. Instead, it’s far more likely that this will be yet another tool human developers will have to create output faster, to create output better, or to otherwise assist human beings in their work.
Tim Sweeney of Epic Games chimed in on the conversation as well.
Ultimately, Sweeney says not to worry: “There’s always a fear that automation will lead companies to make the same old products while employing fewer people to do it,” Sweeney wrote in a follow-up post on X. “But competition will ultimately lead to companies producing the best work they’re capable of given the new tools, and that tends to mean more jobs.”
And Carmack closed with this: “Will there be more or less game developer jobs? That is an open question. It could go the way of farming, where labor-saving technology allow a tiny fraction of the previous workforce to satisfy everyone, or it could be like social media, where creative entrepreneurship has flourished at many different scales. Regardless, “don’t use power tools because they take people’s jobs” is not a winning strategy.”
We have no choice but to progress, in other words. Adapt or die. All the clichés. But we should also have our eyes open to the opportunities AI could create in the gaming industry and others, rather than wallow in doom and gloom.
Filed Under: ai, generative ai, john carmack, llms, tim sweeney, video games
Companies: microsoft
Comments on “John Carmack: Calm Down About Video Game AI Tech Demos, Folks”
Ah, yes, the living embodiment of cyberspace who intentionally wears a meatsuit to contain his otherwise limitless power.
Re:
He’s also a cunt.
John Carmack is headlining a ‘toxic and proud’ sci-fi convention that rails against ‘woke propaganda
Re: Re:
Carmack was at that alt-right “BasedCon” and he’s also a DOGE fan.
Re: Re: Re:
Who would have thought a guy who hopped into bed with Palmer Lucky then Zuckerberg and has always been about the tech rather than storytelling, art or humanity would be a bumhole?
no
nobody ever should be tolerating software capable of hallucinations.
This is the very core of where ai is never acceptable. Nvidia ai frames are a perfect example and where’s addressing the brain drain of ai?
There’s an immediate loss of culture
Re:
We “tolerate” software with bugs, in the sense that we expect them to be fixed by the developers. How would hallucinations be any different?
And sure, if you only focus on the negative aspects, it’s easy to decry AI as the cause of culture loss.
However, AI is also used to modernize old pieces of culture.
As for the brain drain argument, I’m sure the same thing has been said about printed books, music, television, video games and the internet.
Re: Re:
Now, when you say “modernize”, is that like videogame remasters that look worse because they look better, or are you referring to something else?
Re: Re: Re:
It could refer to any number of things, including colorized films, “A.I. up-sampling”, special effects, optical character recognition, machine translation…
(Some of these things are widely hated, I know.)
Re: Re: Re:
Of course I can’t find it again, but I remember reading a while back that AI was being used to restore very old photographs and film reels.
Now, the quality of said restoration is debatable, but OP claiming that AI inherently leads to a loss of culture is bogus in my mind.
Re: Re: Re:2
Yeah, boutique anime company Discotek Media has used the AstroRes upscale process/device/whatever several times to help them remaster or restore footage that couldn’t be done any other way. That process was almost used for the remaster of Project A-Ko until the long-lost masters for that film were found pretty much by accident.
Re: Re: Re:
There’s also nVidia’s “RTX Remix” tool, which, while not fully AI, employs AI to update textures and effects in games. So far, the only prominent project is “Half-Life 2 RTX”, but we’ll likely see more down the line.
Re: Re:
Hallucinations are different in the sense that they’re not directly fixable by developers.
Think of the old Excel bug that turned out to be a bug in the Pentium floating point logic. Entering a specific calculation in an Excel spreadsheet would result in a verifiably incorrect result that could be replicated by anyone. This meant that eventually, it was identified and fixed.
Now with AI, such bugs happen a certain percentage of the time, and we call them “hallucinations”. But re-running the same code and context won’t result in the same result, because the logic relies on randomness at its core. And there’s usually no way for those relying on the model to step through the model’s logic or tweak the results without first identifying the core issue and then building a new model that avoids that issue (good luck identifying this when you can’t replicate the process or reliably examine the weights).
So what “AI” is, is essentially an Excel spreadsheet where any one of the calculations may be wrong, but there’s no way to know how many or which ones, and the errors may or may not be different when you try and replicate the problem. And this randomness is a feature baked into the design.
So sure, if you’re making a movie, and you want to generate a spreadsheet with believable looking content, that someone uses in it but doesn’t depend on for actual results, AI is great. But you’re never going to deploy such a spreadsheet to all employees in an accounting department.
Re: Re: Re:
Machine learning, and most other forms of “A.I.”, do not generally rely on randomness. Give them the same database of “weights” or whatever, and the same input, and they’ll product the same output.
Anyway, I think people are talking about two different things here. Some people are talking about semi-automatically generating game content to be published; normal software testing processes are the way to catch problems here. Others are talking about games that generate content, live, for each player, which indeed makes testing impractical.
Re:
Nvidia’s frame generation seems to have the errors down to a point I probably won’t notice them while playing a game. I know I’ve been happy with games that had worse graphical glitches.
If I get a compatible GPU, the thing that will make me hesitate to turn on frame gen is input latency. I’ll look more into that if using frame gen becomes an option for me.
Re:
What does this mean? Isn’t all fiction, including games, basically “hallucination”?
Re: Re:
In this context, they are referring to AIs inventing things whole cloth. Instead of saying, “I don’t know”, they manufacture answers.
One example is the idiot lawyers that had an AI write a brief to SCROTUS. It referenced case law that didn’t exist. It just invented it on the spot to support it’s writing.
That’s hallucination for AIs, so no.. not the same as fiction being created from someone’s imagination.
The key difference is one is presented as fiction, one is not.
Re: Re: Re:
Yeah, I understand that, but I don’t see how it’s relevant in this context. The “key difference” you mention is not relevant in the context of a computer game, because nobody’s presenting it as reality or mistaking it for such.
The top-level comment was saying (assuming “ai” was a misrepresentation of “A.I.” rather than the Japanese word for love—which would make it an even more cynical comment) that hallucination is what makes the technology unacceptable for games. While I agree it would matter, a lot, for a court document, I don’t see why it matters at all for games.
Re: Re: Re:2
A hallucination in a video game could present itself as a nonsensical object in a given setting, or, depending on how much AI is used, dialogue unsuited for a specific scene. It doesn’t have to be something that clashes with reality.
Re: Re: Re:3
Sure, but games have been published with nonsensical text before. Even high-profile games, like the original Super Mario Bros. 3 saying “Miss twice and your out!”, or the fairy who offers to “sooth your wounds” in Zelda: A Link to the Past. Role-playing games were infamous for bad translations, to the extent that fans have re-translated many of them.
Proper copy-editing and play-testing will catch stuff like that. It’s not like the A.I. is making up the game on the fly. (Or, I suppose it could be, but auto-generated levels and such aren’t new, and fuckups happen.) None of this stuff is terrible enough to say A.I. is “never acceptable” for games, any more than translation is never acceptable.
Re: Re: Re:4
A translation error is not even close to comparable with an AI attempting to manufacture something.
And when you’re making something as complicated as software, it’s usually best to actually try to do it right as best you can the first time than haphazardly slap everything together and then “fix it in post.”
Re: Re: Re:5
Why? Many of the bad translations are manufacturing meanings that weren’t there originally. Does it matter whether Ted Woolsey or ChatGPT is to blame?
I agree, but basically nobody does that. And I don’t think we’re talking about auto-generating the software, either; just the “assets”.
While I agree that the subject is more more complicated and nuanced than all the doomsayers would have us believe, there are still plenty of people out there looking to make a quick buck and use AI to churn out large volumes of low quality slop. It’s already happening with books and music, so it’s only a matter of time before it hits the gaming market as well.
So even if more reputable companies use it solely as a tool in their toolkit, those titles are still going to be harder to find in an endless sea of crap.
Re:
To be fair to AI, that’s already the case now. Spending just a few minutes looking through the worst-rated games on Steam, you’ll find a cornucopia of asset flips and ripoffs.
AI will increase the amount of slop, sure, but so far, I haven’t seen any evidence that people are having trouble finding the quality products.
Re: Re:
Using an AI might require more work than some of the lower effort asset flips.
Re:
That’s basically Sturgeon’s law: “It came to [Theodore Sturgeon] that [science fiction] is indeed ninety-percent crud, but that also – Eureka! – ninety-percent of everything is crud. All things – cars, books, cheeses, hairstyles, people, and pins are, to the expert and discerning eye, crud, except for the acceptable tithe which we each happen to like.”
All of these people freaking out about “A.I.” makes me think of Sousa freaking out about recorded music: “These talking machines are going to ruin the artistic development of music in this country. When I was a boy… in front of every house in the summer evenings, you would find young people together singing the songs of the day or old songs. Today you hear these infernal machines going night and day. We will not have a vocal cord left. The vocal cord will be eliminated by a process of evolution, as was the tail of man when he came from the ape.”
Later, the film and music industries freaked out about VCRs, CD ripping, Napster and other forms of copyright-indifferent Internet transmissions… and yet there always seem to be more music, films, and television shows than ever.
If you actually monitor what the “scene” release groups are putting out, especially for films, it’s surprising how much terrible stuff exists. I’ve rarely seen an IMDb rating below 5 for anything I’ve heard of, but when I search the new scene releases, there’s actually a fair bit in that range. People don’t normally hear about those.
More relevantly in this context, Carmack authored a huge number of games most people have never heard of. Co-worker John Romero once wrote “DOOM was the 90th game I had created”, and it was probably similar for Carmack. That’s not to say those early games were crap; just that, while it’s not hard to churn out games—and I suppose A.I. can do that, or will be able to—it takes a lot of practice to create one of cultural importance.
Wait!
You mean you aren’t doing your chariot wheel maintenance on a regular schedule?!?
But, think of the horses!
Re:
My buggy whip business is being destroyed by AI.
Yes, if the line is going up now, that surely must mean it will continue to go up at the same rate.
Re:
Suggesting that progress will continue != suggesting that progress will continue at the same rate.
Re: Re:
How much stock do you have in Nvidia and other tech corpos that you feel the need to reply-guy everyone rightfully talking shit about AI?
Re: Re: Re:
Oh sweetie, does it make you mad that not everyone joins your club for the technologically pessimistic?
Re:
Well, no; that’s why it was described as only “nearly” inevitable. Humanity’s had a long history of freaking out about technological advancements, the Luddites being probably the best known example—but hardly the first or last.
Most of these freak-outs turn out to be temporary inconveniences to parts of society, while benefitting people overall. The printing press put most scribes out of business. Ubiquitous electronic computers put typists out of business. Probably someone was inconvenienced by photocopiers. Computer games were bad for the makers of board games. It could be wildly different this time, but it’s probably not.
I still remain severely unconvinced that “generative AI” isn’t going to hit some kind of wall within a year or two. There’s not enough data and power and even increasing that apparently has diminishing returns, and even distilled models have major issues.
I wasn’t exactly freaking out about the Quake “demo” either, I mean, it runs at a comically low framerate and the environment changes completely just by turning around. There are fundamental issues with it that I can’t see anyone fixing anytime soon.
Re:
I’m mostly thinking the same.
When an AI is able to produce an entire game on its own, I doubt it’s going to be something like this that generates the game frame by frame in real time.
An AI that generates a game that the players can download onto their device seems like it will be cheaper to run and easier to advertise because the game isn’t different for every player.
It’s going to be used to shift even more work to underpaid and overworked contractors without strong labor protections, the likes of which a lot of game devs are fighting for right now. That’s the core goal.
Re:
Laughs in System Administrator
We’ve needed unionization for decades, but nope. Just more salaried positions that require insane hours, pay rates that only partly keep up with the cost of living and did I mention the fucking hours?…weeeeeee
John Carmack showed his whole ass to be a libertarian whackadoo a while back. Tim Sweeney is a particular flavor of corporate jackass. I honestly wouldn’t take their AI boosterism seriously.
That’s a bit black and white. There’s an in between, where it simply takes way less people. A lot of industries employ way less people than they used to. You already see this in say, factories that are highly automated with robots. Yes, there are still jobs (someone has to design the robots, and someone has to maintain them, after all). But those roles are a) higher skill and b) fewer, than doing it by hand was. You’re also seeing downsizing in e.g. a lot of rote code-writing jobs.
It can be both, as many past technologies have been.
You have to be very very careful with that analogy, because the types of labor are different. To quote economist Brad Delong: labor is actually six dimensions: (1) backs, (2) fingers, (3) brains as routine cybernetic controls for mechanisms, (4) brains as routine cybernetic mechanisms for accounting operations, (5) smiles, and (6) creative ideas. (1) has been–slowly–becoming less valuable since the domestication of the horse. (2) has been–more rapidly–becoming less valuable since the start of the British Industrial Revolution. But diminished value in (1) and (2) has, so far, always, within a generation, been offset by greatly increased value in (3), (4), (5), and (6) as capital complements those other dimensions of labor. Now (3) is falling to the robots and (4) is falling to the ‘bots–leaving us with only (5) and (6). This is qualitatively different from previous episodes of technological change. Will it have different consequences for overall capital-labor complementarity?). (He has some other posts that go into more detail, that I would recommend)
Even in the short term, those still caused a lot avoidable of pain. There are ways we can make the technologic switch, without creating so much suffering. Creative destruction doesn’t need to ruin lives.
Re:
I have to wonder how much value there really is in number 5. Think how easily most people have accepted self-checkouts.
As for 6, there can only be so many people working in roles that rely on it.
This won’t be used to create better output for humans, it will be uaed to drive down wages for artists, coders and composers while being entirely dependant on their output which will continue to be stripmined from the internet without consent. It will be used to cut corners on big projects, especially in marketing but the bulk of the work will be done by artists who work just as hard for a third less pay because ‘the computer can do their job’ according to someone with an MBA who didn’t build or create anything and would fire them all if he genuinely believed the output was reliable enough to sell.
At some point we’re going to have to come to terms with the fact that full employment for all is no longer a realistic strategy.
Smart countries and economies are already looking at and testing alternative models but I fear the USA, because of its unique combination of ‘work ethic’ and innovation, is going to be very badly prepared to handle it.
Re:
Every pilot program for universal basic income in the U.S. has always turned out positive. The only arguments against it are about “how are we gonna pay for it” (by taxing billionaires into becoming mere millionaires) and “what about people who don’t wanna work” (they deserve to live, too).
I will say that if UBI is going to be effective, it should come with a strong social safety net (e.g., free healthcare, affordable housing/rent controls) and the idea of a maximum wage (so executives don’t keep walking away with millions of dollars for doing less work than the people on UBI). I mean, if the predicted AI revolution comes around and puts a lot of people out of jobs, they’re all going to need that safety net sooner or later to keep themselves from becoming homeless.
No means testing. No “this only goes to the worthy”. It has to be for everyone or else it’s pointless.
Re: Re:
Doesn’t “no means testing” kind of conflict with that?
Anyway, it’s an easy thing to write, but how are you gonna do it? Your phrasing suggests a wealth tax. Be specific, please. Some countries have tried and abandoned that; why should we expect the U.S. do better? The only such thing they’ve ever done, to my knowledge, is (“real”) property tax.
Anyway, there’s little hope of the U.S. being a leader here. If they get basic income, they’ll probably be copying some country in Europe. As to drew’s statement that “full employment for all is no longer a realistic strategy”—it was never a realistic strategy, and people just never had the courage to admit it. We just kind of ignored the people living on the streets, or in poverty. Maybe called them “lazy”, “unskilled”, or “welfare queens”, if we thought about them at all.
Re: Re: Re:
Not really. Millionaires could apply for UBI if they wanted to, but it’s not like they can’t afford to live without it. For a lot of people, UBI would be a massive lifeline that could help raise them out of poverty.
Every dollar that someone worth at least 100 million USD makes should be taxed 100%. The existence of billionaires is a failure of morals and policy. What good does their wealth do when it’s hoarded away by a bunch of rich motherfuckers who are so afraid of the rest of the world that they hire small armies as bodyguards and build doomsday bunkers beneath their mansions?
Re: Re: Re:2
So I guess you’re just saying “no means testing”… to apply for and receive UBI. But your taxation system is basically “means testing”.
Define “worth”. Seriously, because if you don’t, or there’s any lack of clarity, we’re gonna be arguing about to no end. We also have to figure out how anyone’s gonna check this. Let’s assume, counter-factually, a taxation agency with proper funding, just to make things easier.
I suppose, à la Clinton, we also have to determine what the meaning of “is” is, that word being kind of implicit in the quoted statement even though you didn’t write it. Does that determination happen once per year, or is money confiscated for every single day someone’s above the limit? How often is one required to re-value illiquid objects such as paintings, and semi-liquid objects such as real property? Will that 100 million about be subject to annual inflation adjustments? Is this a replacement for income tax or something that would operate in parallel?
This gets complicated, too, because “hoarded away” is not really an accurate statement. Few people have 100 million U.S. dollars sitting in a checking account. That money’s invested, in which case, public policy tends to assume it’s doing some good (whether or not it’s true; the people pushing these policies are not free of bias).
Were everyone just “hoarding”, nobody would be able to get a mortgage, for example. If Zuckerberg, Bezos, or Musk pull large amounts of money from their companies—to pay wealth taxes, to stay under its limit, or whatever—those companies would likely have to lay off people (because paying the executives less is unthinkable, and is not required by your plan).
Maybe that’d be a better world, but it’s not obvious.
Re: Re:
Means testing only exists to be a way to make it easier for conservatives to unravel a public service or benefit. People love things that are for everyone and will fight to save them but a lot of people don’t love things that they don’t qualify for, but poor people do and lots want them destroyed out of spite.
It reminds me why Andrew Yang’s take on UBI was so damned insidious and vile. ‘You can have x dollars a month, but only if you opt out of previously universal welfare programs.’ Like his third party, it’s pretty easy to see what he’s doing.
Re: Re: Re:
Why are you suggesting this is “vile”? It’s a pretty standard argument for Universal Basic Income. By making sure everyone has enough to live, we do away with the overhead of administering dozens of programs that need to determine it. Such programs usually have slightly different rules and need to be applied for separately, and the piecemeal design tends to lead to weird results—like millionaires qualifying for food banks, because they consider income but ignore assets such as McMansions.
Re: Re: Re:2
Because it will lead to the death of essential programs designed to help the hungry, disabled and poor entirely and I think we all known full well it won’t match inflation or the value of what it’s supposedly replacing, and a lot more than just food stamps will be rolled in. His version of UBI was not going to be enough to live on, long term. I am pro UBI but when the UBI is designed and pitchrd by a neoliberal tech bro there is zero percent chance it will meet basic needs and will not result in austerity ‘savings’ that take more than they give to those who need it.
Re: Re: Re:3
Okay, I haven’t read about that specific version…
…but I see no particular reason to think Yang’s version will be the one adopted. In general, tax, welfare, and Social Security programs in the U.S. are fully adjusted for inflation, and most basic income concepts call for the same. And I’ve been surprised at how much money they tend to promise; usually way above the poverty line, that line usually being quite a bit higher than what I actually spend.
People do make the mistake of treating inflation-adjusted dollar amounts as real, and even calling them such. The person experiencing exactly-average inflation is as real as the “average pilot”, which is to say they’re nothing more than a mathematical construction.
I think it would make more sense to be worried about the existing social programs not being property adjusted for inflation. That’s dozens of programs that most people are not paying attention to, and some politicians and members of the public are always wanting to cut (based on ideas of “welfare queens” and such). Especially with the current government.
It’s just like we’re having the Luddite situation again:
https://www.flyingpenguin.com/?p=28925
As a general rule, anything John Carmack says can be disregarded. And when Tim Sweeney is backing him up it’s even worse.
Re:
Why?
Re:
You sound very jealous. Are you John Romero by any chance and still salty over Daikatana tanking so bad?
Re:
Yep. He went to the alt-right “BasedCon” a while back, and is a DOGE fan.
https://www.resetera.com/threads/john-carmack-doge-fan.1112796/
People were complaining 20 years ago that video games, because of the internet, would sacrifice offline content because “online” was the word.
Today, it’s very hard to find a split-screen game, but still nobody thinks games were so much better 20 years ago (except, many for the file size, like Oblivion that was stored in a CD-ROM, and now the Remastered on about 200 CDs).
Re:
Plenty of folks actually do think games were better 20 years ago. And they aren’t necessarily right, but axing things like splitscreen and cheats for online play and DLC/microtransactions is not an unvarnished good.
If you think the game industry can be trusted to use this tech without axing labor and/or reducing quality, I have a bridge in London to sell you. AAA gaming already squeezes its customers and audience as much as it can, and “AI” doesn’t complain about crunch, so…
I’m unironically entering my “i want shorter games with worse graphics made by people who are paid more to work less and i’m not kidding” phase with this shit.
Re:
“AAA” gaming isn’t AAA by any stretch. it’s a brand-name like pretending designer handbags are SOOOO much better than other bags etc.
Most “AAA” games are boring snoozefests or repeats of previous games but with palette swaps etc.
Yes, yes it could
“It could go the way of farming, where labor-saving technology allow a tiny fraction of the previous workforce to satisfy everyone”
I’d say using factory farms as an argument in favor of AI is not a winning strategy, either.
To the slaughterhouse!
In quake dad’s world, we’d still be developing via punch cards and “games” would be restricted to 512bytes maximum.
This sounds disturbingly familiar…
Please go watch Extra History’s “The Cotton Gin – Seeds of a Lie”.
Sure thing
mega game exec with shallow morals: I’m sure AI is actually going to create even more middle class jobs, right?
Another game exec with even shallower morals: yeah, just like he said! if you’d quit worrying and learn to work with AI then there will definitely be even more middle class job opportunities
The buggy whip example is alwas funny except cars fucking suck.
Re:
Yeah, this actually makes me wonder how many of the people who are hyping up AI and raging against regulations on it today would, back then, be part of the car industry lobby that caused the U.S. to become so car-centric to our detriment.
It's like NFTs
Who’s asking for this. Who wants it.
The consumers, or the corpos?
Re:
Probably the VCs who keep pumping a lot of money into it, thinking that their investment will pay off big. At some point, the growth in the number of people using AI won’t be enough to satisfy the VCs. All that money they are pumping in will dry up.
Then things get messy. Last I heard, OpenAI had 4 million in revenue and 9 million in expenses in 2024. With that loss expected to increase this year. Every person using it, even those who are paying, is losing money.
Imagine trying to make that profitable.
This comment has been flagged by the community. Click here to show it.
I'm interested in this post and explore my business with you.
Custom packaging for small businesses and big brands is a necessity. Giving away your products in a boring and plain box is a thing of the past. Consumers have gone extremely choosy when it comes to deciding what’s best for them. Even a low price strategy and a high quality product aren’t going to work well. It all comes down to the presentation, the value, and the brand reputation.
Custom packaging is a cost-effective and trusted way to gain customers` attention, assist them in making a purchase, and create a long-lasting impression. However, it’s not easy to win the hearts of the customers and gain a competitive edge over the big giants.
https://flickpack.com/what-is-custom-packaging/