My tepid take on the AI alarmism, particularly re. art, is that commercial art has always served as a form of conspicuous consumption, i.e. as a way for wealthy individuals and institutions to signal status. This was very much the case in the Renaissance: many of the greatest and most costly works of art ever were commissioned by the church as an elaborate flex, and meanwhile wealthy and powerful bankers etc. were funding artists and scholars not just because they loved art and learning, but also in order to one-up each other and show that they could. Being a patron of the arts was high-status because art was costly, and it was costly because it required enormous amounts of human labour and talent.
On the other hand, AI art in any kind of commercial project is already seen as a mark of extremely low status, which signals that your product (whatever it may be) is of very low quality, because if it was high quality, wouldn't you go to the trouble of paying for real art? Even if people hadn't already formed a very hostile anti-AI consensus on basically visceral grounds, which pretty much guarantees that use of AI art commercially will hurt sales, AI would still end up being seen as low status simply because it's so cheap.
It's like with lab-grown diamonds vs mined diamonds - consumers actively prefer the mined diamonds not in spite of them being more expensive, but BECAUSE they are more expensive, and they're more expensive because they required backbreaking labour to extract. On some level, your average western diamond-buying normie likes the fact that children die digging these things up.
Anyway, a very good post all round. I like the cinema comparison, and share the feeling of cautious optimism for the future. Smaller, younger studios are doing incredible things, and it's great to see. You could not pay me to play a Ubisoft game, but on the other hand you couldn't tear me away from KCD2.
Great points. To me, one of the most philosophically interesting parts of the debate centers on AI's mixed role in democratizing the production of media. It's kind of like the impact that woodcuts and lithography had on painting — once any Italian burgher could afford a beautiful print, everybody's attitudes toward the medium had to evolve. The difficulty of (re)producing the artwork ceased to be its primary source of perceived value, in other words. Next week's follow-up will be about how I think game design fits into these calculations.
I resonate with your diamond analogy, too! I got my wife a lab-grown diamond because I could afford twice the carat weight at my budget (back when I had a salary, of course), and I'm glad I did — she gets way more compliments on that diamond than she'd get on a rock half its weight, no matter how much blood was spilled to produce the latter.
Finally, I guess I'm going to have to quit procrastinating and play KCD2 already.
Speaking of analogies - one thing that gets lost in the discourse about whether AI is a good thing or bad thing is that it doesn't have to be completely eliminated Bulterian Jihad style OR completely replace all human input in every realm imaginable. In fact, plenty of stuff in the gaming sphere has ALREADY been taken over by algorithms - SpeedTree and procedural generation as the good and [mostly] bad examples. It's probably perfectly fine for the thousands and thousands of random trees filling out a forest in a bunch of games to get cranked out according to machine-generated patterns - no one's losing a good job or an opportunity for creative expression there. By the same token, plenty of games of the past few decades have squandered promising art, gameplay, and story on worlds that feel hollow and meaningless because making a 'roguelite' with a handful of room shapes slotting into each other arbitrarily is easier than getting out the grid paper.
Oh, right, my analogy. I think AI [or what we now call AI in terms of content generation] will take a few years to settle into a mostly inoffensive role in filling out background fluff that doesn't need to be examined thoroughly or, more importantly, even necessarily be rendered completely, because it can create a convincing periphery while sparing artists a whole lot of mostly pointless work. It'll be like wallpaper - a cheap, quick solution to making a wall not look gross or blank, which isn't as good or appealing as paint or frescoes or molding but is good enough to sell the apartment on. At least, I hope that'll happen as its profound failings sink in. But given the amount of money people have dumped into that pit, I wouldn't be surprised if more gets thrown in for a while.
I think you're probably right. A lot of folks don't realize how much mind-numbingly repetitive busywork is being replaced by algorithms and indeed by gen-AI. There's a persuasive case to be made for AI's potential in fighting back against overwork and crunch in high-level game design. SpeedTree is an excellent example — modelling, texturing, and shading trees from scratch is a goddamn nightmare, and I dread to imagine having to do it as a nine-to-five.
But you're also right about the foibles of the roguelite proc-gen obsession. Perhaps the worst-kept secret in game development is that procedurally generating rooms and levels is way, way faster and cheaper than designing bespoke environments from scratch. Gen-AI might just represent a happy middle-ground in which human designers manage the broad strokes and leave the minute details to the machines, but I suspect the industry will have to banish the AI hype-mongers before it can start to make serious moves in that direction.
I'd been hoping the AI tech bros would be slapped down as quickly and as humiliatingly as the NFT peddlers (and yes, there is a massive overlap between these two groups). Sadly that hasn't come to pass.
There are plenty of legitimate use cases for AI in many fields but the relentless focus on deploying it in the creative arts is, IMO, a giant bummer.
I saw a Reddit post yesterday with some jerk promoting his 'AI-Powered Newsletter Generator for Creators' which promised to write your newsletter for you to 'save time'.
Well said. I guess the essential difference is that NFTs never came anywhere close to gen-AI's level of consumer fascination. That said, it's encouraging to see more and more folks drawing lines in the sand.
And in any case, we'll stay at least one step ahead of the machines as long as we keep writing newsletters about arthouse FPS games and about how God is an SF2 cabinet. Don't give up the fight!
I feel like we've been in the phase of indie productions eating major publishers' lunches for a long while now, and suspect most of the illusion that the big budget model still works has been propped up by executives and studio owners desperate to keep their kingdoms from collapsing. I can't even think of a 'safe bet' big release that actually delivered in recent memory aside from Elden Ring. Remember when Ubisoft claimed Skull and Bones was QUADRUPLE A game? That kind of doubling down [or quadrupling down] doesn't happen when you're not drenched in flop sweat. And I won't miss any of them once they're gone - its been proven over and over now that the indies can completely exceed any expectations that people have for the legacy franchises or genres they draw from.
While I agree with most of your sentiment, I want to argue a little bit with your dismissive tone towards the preeminent Zoomer-gen games, and especially in comparing them directly to the dense narratives of Disco Elysium and BG3. I honestly feel extremely heartened when I remember that three of the most influential games for the up-and-coming generation are Minecraft, Fortnite, and [though you didn't mention it] Roblox, not because I think those games are GOOD, but because I see the appeal of all of them, specifically that they all, at their highest level, reward communal interaction and creative expression. The way I see it, there's an entire generation of children [who are swiftly becoming not just children] that have been raised in a sort of digital prison, and those worlds are their parallel realities where they're able to interact with peers, build treehouses, and run around being a menace like every prior generation was allowed to do outside. I'm sure they'll have their own versions of deep, thoughtful narratives, once they're old enough to reflect on life experiences, but for the time being, it shouldn't be a surprise that they're boosting the numbers of sandboxes that most adults find lacking. Now, the fact that desperate suits are using that to milk them for credit card charges is less heartwarming, but let he who has not stolen quarters from Mom's purse for the arcade throw the first shade.
Man, I had totally forgotten about that goofy "quadruple A" gambit! What bittersweet irony that Ubisoft's growth strategy is now so reliably absurd as to be passé.
Great point about the social/creative value of what we may as well call the Big Three. I wholeheartedly agree with that much — I'd even go a step further and say that Minecraft is a great game on the merits, and if I sound at all dismissive, it's because I still feel aggrieved by its transformation from plucky indie darling to exploitative corporate mega-franchise. It ceased to exist primarily for the purpose of enriching our creativity and became a vehicle primarily for enriching shareholders. My pals and I used to spend all day messing around on private Minecraft servers just for the sake of being creative together, but that was [*checks watch and vomits*] about fifteen years ago.
In the time since, I feel like it's become an instrument of the very digital prison you're talking about. I have similar impressions about Fortnite and ROBLOX, which I think have undergone comparable transformations in character. All certainly still reward communal interaction and creative expression, but I reckon they all did better jobs at each before becoming cash-cows for publicly traded colossi.
I guess my point is that we can and should have it both ways. Not "Disco Elysium's narrative with Minecraft's gameplay," mind, but a robust market of affordable games such that we can freely choose between dense narrative experiences and relaxing sandboxes without having to put up with the foibles of corporate enshittification. Fingers crossed.
Thanks so much for reading and for your thoughts! You gave me plenty to think about for Part II.
My tepid take on the AI alarmism, particularly re. art, is that commercial art has always served as a form of conspicuous consumption, i.e. as a way for wealthy individuals and institutions to signal status. This was very much the case in the Renaissance: many of the greatest and most costly works of art ever were commissioned by the church as an elaborate flex, and meanwhile wealthy and powerful bankers etc. were funding artists and scholars not just because they loved art and learning, but also in order to one-up each other and show that they could. Being a patron of the arts was high-status because art was costly, and it was costly because it required enormous amounts of human labour and talent.
On the other hand, AI art in any kind of commercial project is already seen as a mark of extremely low status, which signals that your product (whatever it may be) is of very low quality, because if it was high quality, wouldn't you go to the trouble of paying for real art? Even if people hadn't already formed a very hostile anti-AI consensus on basically visceral grounds, which pretty much guarantees that use of AI art commercially will hurt sales, AI would still end up being seen as low status simply because it's so cheap.
It's like with lab-grown diamonds vs mined diamonds - consumers actively prefer the mined diamonds not in spite of them being more expensive, but BECAUSE they are more expensive, and they're more expensive because they required backbreaking labour to extract. On some level, your average western diamond-buying normie likes the fact that children die digging these things up.
Anyway, a very good post all round. I like the cinema comparison, and share the feeling of cautious optimism for the future. Smaller, younger studios are doing incredible things, and it's great to see. You could not pay me to play a Ubisoft game, but on the other hand you couldn't tear me away from KCD2.
Great points. To me, one of the most philosophically interesting parts of the debate centers on AI's mixed role in democratizing the production of media. It's kind of like the impact that woodcuts and lithography had on painting — once any Italian burgher could afford a beautiful print, everybody's attitudes toward the medium had to evolve. The difficulty of (re)producing the artwork ceased to be its primary source of perceived value, in other words. Next week's follow-up will be about how I think game design fits into these calculations.
I resonate with your diamond analogy, too! I got my wife a lab-grown diamond because I could afford twice the carat weight at my budget (back when I had a salary, of course), and I'm glad I did — she gets way more compliments on that diamond than she'd get on a rock half its weight, no matter how much blood was spilled to produce the latter.
Finally, I guess I'm going to have to quit procrastinating and play KCD2 already.
Speaking of analogies - one thing that gets lost in the discourse about whether AI is a good thing or bad thing is that it doesn't have to be completely eliminated Bulterian Jihad style OR completely replace all human input in every realm imaginable. In fact, plenty of stuff in the gaming sphere has ALREADY been taken over by algorithms - SpeedTree and procedural generation as the good and [mostly] bad examples. It's probably perfectly fine for the thousands and thousands of random trees filling out a forest in a bunch of games to get cranked out according to machine-generated patterns - no one's losing a good job or an opportunity for creative expression there. By the same token, plenty of games of the past few decades have squandered promising art, gameplay, and story on worlds that feel hollow and meaningless because making a 'roguelite' with a handful of room shapes slotting into each other arbitrarily is easier than getting out the grid paper.
Oh, right, my analogy. I think AI [or what we now call AI in terms of content generation] will take a few years to settle into a mostly inoffensive role in filling out background fluff that doesn't need to be examined thoroughly or, more importantly, even necessarily be rendered completely, because it can create a convincing periphery while sparing artists a whole lot of mostly pointless work. It'll be like wallpaper - a cheap, quick solution to making a wall not look gross or blank, which isn't as good or appealing as paint or frescoes or molding but is good enough to sell the apartment on. At least, I hope that'll happen as its profound failings sink in. But given the amount of money people have dumped into that pit, I wouldn't be surprised if more gets thrown in for a while.
I think you're probably right. A lot of folks don't realize how much mind-numbingly repetitive busywork is being replaced by algorithms and indeed by gen-AI. There's a persuasive case to be made for AI's potential in fighting back against overwork and crunch in high-level game design. SpeedTree is an excellent example — modelling, texturing, and shading trees from scratch is a goddamn nightmare, and I dread to imagine having to do it as a nine-to-five.
But you're also right about the foibles of the roguelite proc-gen obsession. Perhaps the worst-kept secret in game development is that procedurally generating rooms and levels is way, way faster and cheaper than designing bespoke environments from scratch. Gen-AI might just represent a happy middle-ground in which human designers manage the broad strokes and leave the minute details to the machines, but I suspect the industry will have to banish the AI hype-mongers before it can start to make serious moves in that direction.
I'd been hoping the AI tech bros would be slapped down as quickly and as humiliatingly as the NFT peddlers (and yes, there is a massive overlap between these two groups). Sadly that hasn't come to pass.
There are plenty of legitimate use cases for AI in many fields but the relentless focus on deploying it in the creative arts is, IMO, a giant bummer.
I saw a Reddit post yesterday with some jerk promoting his 'AI-Powered Newsletter Generator for Creators' which promised to write your newsletter for you to 'save time'.
Handy, no?
Well said. I guess the essential difference is that NFTs never came anywhere close to gen-AI's level of consumer fascination. That said, it's encouraging to see more and more folks drawing lines in the sand.
And in any case, we'll stay at least one step ahead of the machines as long as we keep writing newsletters about arthouse FPS games and about how God is an SF2 cabinet. Don't give up the fight!
I feel like we've been in the phase of indie productions eating major publishers' lunches for a long while now, and suspect most of the illusion that the big budget model still works has been propped up by executives and studio owners desperate to keep their kingdoms from collapsing. I can't even think of a 'safe bet' big release that actually delivered in recent memory aside from Elden Ring. Remember when Ubisoft claimed Skull and Bones was QUADRUPLE A game? That kind of doubling down [or quadrupling down] doesn't happen when you're not drenched in flop sweat. And I won't miss any of them once they're gone - its been proven over and over now that the indies can completely exceed any expectations that people have for the legacy franchises or genres they draw from.
While I agree with most of your sentiment, I want to argue a little bit with your dismissive tone towards the preeminent Zoomer-gen games, and especially in comparing them directly to the dense narratives of Disco Elysium and BG3. I honestly feel extremely heartened when I remember that three of the most influential games for the up-and-coming generation are Minecraft, Fortnite, and [though you didn't mention it] Roblox, not because I think those games are GOOD, but because I see the appeal of all of them, specifically that they all, at their highest level, reward communal interaction and creative expression. The way I see it, there's an entire generation of children [who are swiftly becoming not just children] that have been raised in a sort of digital prison, and those worlds are their parallel realities where they're able to interact with peers, build treehouses, and run around being a menace like every prior generation was allowed to do outside. I'm sure they'll have their own versions of deep, thoughtful narratives, once they're old enough to reflect on life experiences, but for the time being, it shouldn't be a surprise that they're boosting the numbers of sandboxes that most adults find lacking. Now, the fact that desperate suits are using that to milk them for credit card charges is less heartwarming, but let he who has not stolen quarters from Mom's purse for the arcade throw the first shade.
Man, I had totally forgotten about that goofy "quadruple A" gambit! What bittersweet irony that Ubisoft's growth strategy is now so reliably absurd as to be passé.
Great point about the social/creative value of what we may as well call the Big Three. I wholeheartedly agree with that much — I'd even go a step further and say that Minecraft is a great game on the merits, and if I sound at all dismissive, it's because I still feel aggrieved by its transformation from plucky indie darling to exploitative corporate mega-franchise. It ceased to exist primarily for the purpose of enriching our creativity and became a vehicle primarily for enriching shareholders. My pals and I used to spend all day messing around on private Minecraft servers just for the sake of being creative together, but that was [*checks watch and vomits*] about fifteen years ago.
In the time since, I feel like it's become an instrument of the very digital prison you're talking about. I have similar impressions about Fortnite and ROBLOX, which I think have undergone comparable transformations in character. All certainly still reward communal interaction and creative expression, but I reckon they all did better jobs at each before becoming cash-cows for publicly traded colossi.
I guess my point is that we can and should have it both ways. Not "Disco Elysium's narrative with Minecraft's gameplay," mind, but a robust market of affordable games such that we can freely choose between dense narrative experiences and relaxing sandboxes without having to put up with the foibles of corporate enshittification. Fingers crossed.
Thanks so much for reading and for your thoughts! You gave me plenty to think about for Part II.