A lot of people in the comments say that the supposedly better examples in the article are still “horrible” because they can tell they’re AI and point to things that the AI did “wrong” and a human artist would do better.
That may be true if you’re working with a top of the line artist, but in my experience the average freelance designer online is significantly worse than AI. I’ve worked with dozens of budget friendly designers on Fiverr and consistently get way worse results than I’ve gotten from AI models. Will a great human designer produce better output than an AI? Maybe, but a small local event or business probably won’t have the budget to hire them or know where to find them. It’s cheaper, faster and easier to use AI to generate something that’s good enough for their purposes.
People also point out all the little flaws in the “good” examples and cite them as obvious tells that they were AI generated. I think it would be interesting to see if people could reliably distinguish between AI generated flyers and human made flyers. Sure, sometimes it’s obvious like when the AI adds those weird, cartoony people. But if the model is told not to create designs like that and stick to the styles used in the second half of the article, I expect people would have a much harder time distinguishing the human designs from the AI designs.
It sure seems like, while LLMs and diffusion models won't completely eradicate the middle class, the bar has now risen significantly to actually contribute in a lot of areas.
For programmers, sure SV startups and actual software-first shops are probably only augments with LLMs and not replacing. But what % of software engineers are employed in such shops? Well, from my graduating class in CS, approximately 30% of grads all went to the same contracting shop whose business model seems to be, take the bottom 30% of each CS class, underpay the fuck out of them, bootcamp them into something vaguely usable + teach the house style of CRUD/LOB app framework, then throw them at clients. And this seems to still be a step above "offshore to barely baseline capable (or worse) body shop for extremely dubious results."
That entire bottom 50% is probably about to be eviscerated. The "offshore + pray" approach is probably going to be completely zapped, much easier to just vibe code for probably better (or at least not worse) results. And the onshore-but-cheaper contracting shops are probably going to need a lot fewer people whose main contribution is translating low-ambiguity tickets into syntactically valid java spring code.
Similarly, for artists, disney is probably not going to use AI in a way that substantially speeds them up because they won't tolerate the quality loss at all. But how many artists work for disney? And how many make anime waifus for the gacha apps? The latter of which I can only assume has already switched substantially to AI assets.
Given their empire these days: a lot. Mostly because outside of Disney there's really not much such thing as "modern salaried artist". You work for any of Disney's 50 subsidiaries, you work at a games studio (because Disney gave up its expansion early in that medium), or you're likley freelancing/commissioning and grabbing whatever gig comes along.
I'm exaggerating, but less than I care to adnit. Disney makes up that much of "normal" full time artistry these days. And it's no secret how little respect artists have had in the US, even before AI.
>And how many make anime waifus for the gacha apps? The latter of which I can only assume has already switched substantially to AI assets.
In America? Less than you think. Asia has plenty of cheaper and more talented artists to exploit. AI adoption is interesting there because it's already such a sweatshop that you need to consider if AI really saves that much time for its cost. Larger studios are dabbling, smaller ones use sparingly.
No, North/South American slop is fully focused on the hyper-casual market. It's fueling thousands of asset flips for quick mobile games or copying indie games into Roblox/Fortnite. That's where the effort (or lack thereof) is going.
ajjenkins · · focus · HN ↗
That may be true if you’re working with a top of the line artist, but in my experience the average freelance designer online is significantly worse than AI. I’ve worked with dozens of budget friendly designers on Fiverr and consistently get way worse results than I’ve gotten from AI models. Will a great human designer produce better output than an AI? Maybe, but a small local event or business probably won’t have the budget to hire them or know where to find them. It’s cheaper, faster and easier to use AI to generate something that’s good enough for their purposes.
People also point out all the little flaws in the “good” examples and cite them as obvious tells that they were AI generated. I think it would be interesting to see if people could reliably distinguish between AI generated flyers and human made flyers. Sure, sometimes it’s obvious like when the AI adds those weird, cartoony people. But if the model is told not to create designs like that and stick to the styles used in the second half of the article, I expect people would have a much harder time distinguishing the human designs from the AI designs.
atomicnumber3 · · focus · HN ↗
For programmers, sure SV startups and actual software-first shops are probably only augments with LLMs and not replacing. But what % of software engineers are employed in such shops? Well, from my graduating class in CS, approximately 30% of grads all went to the same contracting shop whose business model seems to be, take the bottom 30% of each CS class, underpay the fuck out of them, bootcamp them into something vaguely usable + teach the house style of CRUD/LOB app framework, then throw them at clients. And this seems to still be a step above "offshore to barely baseline capable (or worse) body shop for extremely dubious results."
That entire bottom 50% is probably about to be eviscerated. The "offshore + pray" approach is probably going to be completely zapped, much easier to just vibe code for probably better (or at least not worse) results. And the onshore-but-cheaper contracting shops are probably going to need a lot fewer people whose main contribution is translating low-ambiguity tickets into syntactically valid java spring code.
Similarly, for artists, disney is probably not going to use AI in a way that substantially speeds them up because they won't tolerate the quality loss at all. But how many artists work for disney? And how many make anime waifus for the gacha apps? The latter of which I can only assume has already switched substantially to AI assets.
johnnyanmac · · focus · HN ↗
Given their empire these days: a lot. Mostly because outside of Disney there's really not much such thing as "modern salaried artist". You work for any of Disney's 50 subsidiaries, you work at a games studio (because Disney gave up its expansion early in that medium), or you're likley freelancing/commissioning and grabbing whatever gig comes along.
I'm exaggerating, but less than I care to adnit. Disney makes up that much of "normal" full time artistry these days. And it's no secret how little respect artists have had in the US, even before AI.
>And how many make anime waifus for the gacha apps? The latter of which I can only assume has already switched substantially to AI assets.
In America? Less than you think. Asia has plenty of cheaper and more talented artists to exploit. AI adoption is interesting there because it's already such a sweatshop that you need to consider if AI really saves that much time for its cost. Larger studios are dabbling, smaller ones use sparingly.
No, North/South American slop is fully focused on the hyper-casual market. It's fueling thousands of asset flips for quick mobile games or copying indie games into Roblox/Fortnite. That's where the effort (or lack thereof) is going.