AI will dominate the animation industry in less than 5 years
nicksaraev.com
nicksaraev.com
I definitely don’t think the author is wrong here. It is very likely AI animation tools will have a huge impact on the industry.
I have friends who work in animation and illustration professionally. Based off their predictions, the truth is likely far more mundane. These AI models will likely just become tools that automate or simplify existing animation workflows for larger projects. This would definitely result in less labor needed for any given project.
For smaller projects. You might see folks use AI for the entire thing
> “the AI art movement is already causing a sizeable reduction in artist commissions, and that effect will soon trickle down into B2B and wider world.”
This line specifically made me lose confidence in the author’s ability to predict the future. Cursory Google searches return nothing. This is also a very difficult thing to measure at a population level. In my friend group of illustrators (which is a very small sample of the population) all of them have used TikTok and seen their commissions drastically increase in the last few months. That doesn’t mean ALL artists are making more money though. Neither does a few mediocre artists losing commissions due to AI either.
But as shown in the video you still need to hand tweak a great deal even if you don’t mess with individual frames. Suppose someone is jumping through a window, you need the sound of breaking glass to sync with the glass being broken.
It’s hard to explain why some things the article suggests are really easy and other stuff is difficult, but context is still a hard problem for AI. https://xkcd.com/1425/
Most or all top anime content is still hand drawn, even the in between frames.
Mind you there is CGI in these animes, but the frames themselves are still hand drawn. How else would you expect them to be done?
There's no magic here.
Cartoons can do it because they just use rotations, translations and skews, anime does not have that luxury. The pro is that, imo, when done right, anime is just beautiful to see, mesmerising even.
It would be neat to be able to have fewer key frames and better in betweening so that more storytellers could have access without a team of CalArts grads. That doesn't seem unreasonable given what we've already seen, but I'm neither an animator nor an AI expert so I'm curious.
That's got nothing to do with AI, it's just economics, and it's been like that for years. Everybody doing animation on the cheap wants to avoid drawing every frame.
Prior to digital animation you used other techniques, like recycling frames or tracing over existing drawings (Disney used to do this all the time in their films).
What digital animation really allowed to happen was 2D puppeteering, where you could rig puppets and animate them in two dimensions without having to draw the key poses or the in-betweens. Now we're at the point where the standard way to do 2D animation is via puppets, rather than any sort of hand-drawn (be it key poses or full tweening)
Not everything comes with a citation. Things first happen, then get studied, even if all the signs are there.
(Also, 'little citation' my ass. The post has links to video animation AI models, studios that already do this thing, papers and relevant technologies, etc. And it's just a post, not an exhausting monography on the subject)
>I have friends who work in animation and illustration professionally. Based off their predictions, the truth is likely far more mundane.
"It's difficult to get a man to understand something when his salary depends on not understanding it" as the saying goes.
The change will not be driven by animators, but by people hiring animators that decide to cut costs by not hiring them.
Subpar results wont be a problem, if they're still good enough. There will be people helping and people touching up the end results, but half or less of what there are now...
I.e.: " These AI models will likely just become tools that automate or simplify existing animation workflows for larger projects. This would definitely result in less labor needed for any given project."
Yes, but much more dramatically so, to the point that there will big loss of jobs in the industry.
I see no reason this couldn’t happen for animation. Studios demand more immersive, interactive experiences that require more and more detail. More animated content is generated. Every marketing team now needs 3D animators to make social media content, etc.
This seems like the logical progression and doesn’t necessarily result in a loss of jobs.
Because software had room to grow into all industries - which huge ROI, applicable everywhere.
Whereas animation has a capped target: it's content, of a specific type even, not a way to help other industries generate profit, like software is.
The industry will still be there. But how much money you can make, and how many people you can afford to have on the payroll, that will diminish.
I think Corridor Crew is showing this effect with virtually every post to their YouTube channel.
I have no doubt that management hopes that they will be able to cut costs through AI! My doubts come from the fact that the people most certain that this possible future will come to pass seem to be armchair experts outside the industry where people in the industry seem mostly excited about new tools and bemused with the IP issues around AI.
It's not like the last 20 years have been free of disruptions from software tools? I have no more of a crystal ball than anyone else, but this is only one of many confident predictions about AI killing human experts that has not yet come to pass. It's certainly possible, but I wouldn't invest based on it.
NeRF lets you pick ANY smooth camera path in a 3D space built up with nothing more than a mobile phone. That's today.
These two phenomena will start to merge and the impact will be as transformative to the software side of media creation as the switch to digital from celluloid was for hardware. There was a time some still remember when not everyone had a camera in their pocket.
Note also, NeRF is not an “AI” model, it is more of fancy compact storage for objects, images or scenes.
The problem for artists is that the vast majority of the skills they learned as artists are useless when it comes to forming prompts. Assuming these models raise their quality and responsiveness a bit, and are surrounded with tools to fix expected problems (anatomy, symmetry, light, etc), I wouldn't expect a professional artist even with an aptitude for the tools to have much better results than an obsessed 14-year-old without a social life who's been playing around with them for a few months.
They're weird in that they in a way turn the process of creation over to the fans. It's like a declarative language for entertainment. While vision and creativity make artists stand out, the real skill for a artist is knowing how to use artists' tools. The real skill for the future painter probably is knowing how to use the AI tools.
People who typeset with metal had very few skills that transferred to InDesign.
Doesn't this artistic vision and style boil down to taste, anyway? Claiming that the skill of the artist and the skill of the editor is identical is obviously wrong.
edit: in a way, making a work that targets a particular audience is an acknowledgement that the public has largely set the vision and style, and the artist is actually adept at working within it.
I am working on a tool to make that happen.
Religious traditions are not sacrosanct, and neither are contemporary civic life traditions. I don’t owe the past carrying on its modern patronage system of copyright maximalism and monopoly ownership of assets to extract rents from.
I think we'll see the opposite, a few elites using AI to generate the content they force everyone else to consume.
Perhaps there will be a black market for AI's, there already is one for illegally trained anime AI's.
I would not be so sure there isn’t a massive amount of potential energy among the public to put video game companies out of work and increase the pool of “essential workers”.
IMO software engineers should optimize for themselves by walking away from industry rather than give it ownership of their intellectual output to be wielded against them after code writing AI is trained against their work.
Y’all thought you were so smart but hubris often comes with the later realization the effort was used against them. If society can generate content without paying someone it will.
I find it odd (but predictable) that people miss this. We’ve seen proliferation of other specialized craft in the past as the tools get better, particularly in programing. And it is not like non-linear systems are that rare among people that are looking into AI and making predictions around it.
Animation is an amazing craft, our culture would for sure benefit from having more of it, and as more and more stuff gets animated, more people are probably gonna get jobs doing it, despite the fact—or even because of the fact—that every individual can produce more of it then before.
What’s likelier is that ML-based tooling becomes a key part of the animation workflow. Used to generate assets, animations, backdrops, characters, etc, which are then combined by an animator/editor. The examples he cites in the article all fit this mold - the anime character generator he cites in the article uses separate models to generate the character then rig it from facial data.
After working in the self-driving car industry, I’m really skeptical of any claim that rapid advances in one modality or task mean we’re “just 3 years away” from all related tasks being done via ML models. Alexnet et al completely revolutionized perception in self driving cars - between 2014-2017 it was really common to hear predictions that we’d have end to end models driving our cars perfectly in “less than 5 years”. That reality never arrived because ML just wasn’t capable of handling more complex tasks the way it could with object detection. Lots of articles similar to this one talking about what we were going to do with all of the out of work truckers and Uber drivers.
And yeah obviously generative art is a different beast than autonomous driving. And I’ve seen examples from WIP text2video models. But I just want to caution that tons of progress in image and text doesn’t necessarily mean we’re just a year or two away from all related tasks being conquered.
especially challenging will be (as it is for self driving) context permanence.
that said, with 3d modelling, rendering etc eating up so much of modern animation costs as it is, it is quite likely that tech as ai art tailored upscalers will merge with technique as dlss to enhance permanence and will help make the cost crash back down to earth, and that will have a big impact as currently only "safer" films get greenlighted due the difficulties from recovering costs.
My point wasn’t that these systems will never work in some capacity - my point is that they’re taking a lot longer to roll out than people predicted, and that the internals of these systems aren’t end-to-end ML models - it’s ML-based perception feeding in to a lot of traditional robotics code, with ML models handling certain prediction tasks in certain spots.
At this point we have a public robotaxi service in one suburban area and two robotaxi services operating under restricted domains in SF that are open to the public behind a waitlist. The cars have problems with rain, fog, or snow, and expanding to a new city still takes a ton of time. Having joined the industry after the hype of 5 years ago, the reality now is nowhere near the vision companies we’re selling back then.
It’s like the future will be everybody living in NvidiaBlackrockDominoes housing, eating TysonIntelJ&J gruel, and the only employer left will be TSMCOnlyFans — which will be a staffing agency for test subjects in GarnierRevlon’s automated anti-aging research facilities.
The only thing I can think of is violent revolution but with the way things are going, the wealthy elite will have a monopoly on violence too.
As for what has historically led to “utopia for the ultra rich” being less-than-infinite in duration… *gestures to wars, petty infighting, boredom, gross incompetence, bad luck*
The current generation of the ultra rich might see that and argue “Aha! But we will have basically used a neural network to generate a society based off of ‘what if Kafka wrote Robocop Star Trek prequel fanfic’! Problem solved!” to which I’d say “cool”
The oligarchs who try to concentrate wealth will still try to scapegoat AI via mass media though, along with Mexicans, boomers, etc. or (historically), Jews.
I work in an animation adjacent field and this is hogwash. What happens when the person responsible for the creative content says “I don’t like the way this part of the sky looks in the background”? Or “The way this character falls down the stairs is too violent and not funny enough”? Who’s going to sit there with the AI and make it do the thing that someone has asked for? It’s not going to be the director or producer because they have too many other things to do. They will delegate that task to someone. Is that person an animator? Does it matter that they’re working with an AI instead of moving keyframes around?
Yes, because prompting an AI requires much less skill than animating it yourself. Do you think "AI prompters" will earn the same amount as animators?
s/much less skill/different stills/
> Do you think "AI prompters" will earn the same amount as animators?
I do, because animation is not inherently valuable. The value they are paid for is solving a problem, the same problem AI prompters will solve. It is totally valid to argue that AI will reduce the total amount of animators, I think that will happen.
Much less skill and different skills. Declaring that it will take the same level of skill (measured in mean amount of training and practice to do the job) in order to make a finished product is an appeal to the law of averages. It does not take the same amount of skill to typeset a book well in InDesign as it did to typeset a book in e.g. the 1930s.
I don’t think that will happen because I believe induced demand is a thing. I think animators will produce more and more efficiently, I also think that amateurs will be animating stuff that they’re currently not animating because it is too hard. I think we will see a huge proliferation in animation as the craft becomes easier, and with the proliferation professionals will not only keep their jobs, but they will actually see their profession expand.
As a comparison, the textile industry still employs millions of experts despite the craft having been automated with the automatic loom as early as the 1780s.
https://ourworldindata.org/grapher/number-of-people-employed...
To continue with the fabric industry. The automatic removed jobs in weaving fabric, but it made fabric cheaper and created many times those jobs in sewing this fabric.
The typical animator doesn't make much. I knew one who worked for a well known movie animation studio. They hire as needed and then let them go easily after the movie is made. He made enough to pay bills, but not much more.
People who write as though not only can they predict the future, but they're so arrogantly certain of it that they write long, scary sounding articles about it.
Guess what? This random guy has no idea what will happen in less than 5 years, no one has any idea. I'd bet AI won't, in fact, dominate the animation industry. Am I certain of it? No, because at least I have some humility.
Anyone who thinks they can predict the future is full of shit.
Well, many people can and have, with reasonable arguments that panned out 20, 30 and even 100 years later. It's not metaphysical prediction, like an oracle, it's just reading trends.
>Anyone who thinks they can predict the future is full of shit.
In any case, we didn't learn anything about what will or will not happen in this comment, or about the feasibility of predicting the future even. But we did learn that you strongly dislike the idea that it's possible.
Also, as if "5 years" is some big leap in predicting an industry change, or technology trend. E.g. I could tell you already in 2007 than in 5 years touch-based smartphones like iPhone will dominate. I could you in 1995 that this "world wide web" thing will go huge.
That kind of prediction is basically what any businessman does when they make an investment in infrastructure or direction that will pay for itself in 5 years or more...
Did you? And more pointedly, any more specific than “huge” and “dominate”?
>any more specific than “huge” and “dominate”
Like? What the list price for AMZ and APPLE stock would be in 2022? Or what color socks would Zuckerberg wear?
Not to mention, that if you dismiss a prediction as "dominate" as vague enough to be easy to come true, then you've already made my point: that's exactly what the TFA predicts.
Your choice is this, or SEO farm bot posts (perhaps ironically).
At first, there was great resistance and many cries of "It's not as good!," which were, of course, true. But it didn't matter. Desktop publishing as it was called (now something your child does in elementary school) was good enough and opened the door to many uses of "the press" which were simply unaffordable before.
The same is coming for media production.
I think people who have seen this process in print and are aware of it have a sharply different view of the potential of this then people who haven't. Seeing the amount of skill, knowledge and practice accumulated over decades by pressmen and prepress/typesetters; all of that is almost worthless now. And by that, I mean it pays nothing, and you're competing with everybody who is "pretty good at Illustrator and InDesign."
It’s like programming, easier it is too program the more creative programs become.
Also in my opinion creativity is just combination on two ideas together. These AI tools allow you to combine different concepts together and see how does it looks.
The issue with articles like the original post is that typically they are written, not by animators or artists, but by data scientists.
This becomes obvious when they approach animation with a lens of realism. In fact, the photograph has not replaced all visual media because artistic expression goes well beyond a facsimile of what we see and experience. The difference between a Degas and a photograph of a ballerina is multifaceted.
Animation is no different, and naturally AI will be a useful tool, but without understanding art in it’s entirety, and without being able to create art oneself, without the aid of a machine will separate the plebs from true artisans.
If AI gets to the point where it can have limitless creativity—then I hate to say it but we’re all fu*ed.
Generated text looks like the person crafting it is having a constant stream of micro-strokes while writing in their third language.
Deep Fakes have a proportion problem. Not to mention little things like skin tone and lighting.
Images look like one of those “one person draws one part, and hands it off blind to someone else” activities.
And animation looks like the characters on screen are also having micro-strokes.
“It will get better,” but will it? Or will we just continue to live in the valley with a ceiling of “it’s good enough”? Getting better will be a huge cost, after all.
NVIDIA seems to have solved that problem: https://the-decoder.com/nvidias-ediffi-is-an-impressive-alte...
The tech is improving extremely rapidly, which makes sense considering how much excitement and interest there is.
jesus this is so stupid on its face its laughable. this is like an artist saying by 2025 that artists will have replaced coders. you have no idea what you’re talking about and how stupid you sound
How many people still code in assembly vs (say) Javascript?
it's relatively easy to label images with text, yet much more complicated to do the same for complicated software systems.
a bigger threat to swe is probably off-shoring, which will probably pick up as a result of remote work popularity and economic belt tightening.
I don't think any of the current AI tech demonstrates any kind of "understanding." The image-generators are closest, but if AI-generated images were code, almost all of them would be unusable.
I think coding jobs are safe until AGI.
That being said, I think copilot and similar can fill in a lot of blanks, because we often solve the same problems repeatedly. (I haven't used them FWIW). They're like a library of solutions, not a brain that invents a new solution.
Honestly I could see an AI replacing many many of our business analysts, artists, ux engineers, and configuration developers.
Kricfalusi (of Ren and Stumpy fame) famously bad-mouthed Tartskovsky's style as lazy and cheap looking. But while it took an army ages to make one season of Ren & Stumpy, Cartoon Network got dozens of episodes of three shows.
... And ultimately, the story, characters, and writing can matter more than the art.
That is a highly subjective conclusion, and worth consideration.
I still remember watching Ren And Stimpy as a kid when it aired, and later Invader Zim and have a way of recalling the skits/gags in a way that I can only due to those shows hook was the art: mainly because how remarkable it looked for it's time. Ren and Stimpy was something like a cross-between Dali and Goya surrealism (specifically the grotesque works when they did close-ups) that to this day stand up; I recall they sold the rights and it was on some other channel when I was in university on cable tv and it looked like a badly made knock-off as a cash grab so after the short lived nostalgia you realized how bad it was because they didn't take the time to do it like in the 90s.
The same can be said about the Nick era plot/art of Vasquez's work with Invader Zim, the art was so unique and it looked like the harshness in his comics and the plots were always weaving between dark and light in such a way that one questions who this show was really written for; the later Netflix stuff (while more advanced in it's appearance) suffered so much, and I felt the sane way about the comics and kind of dissapointed by his claim to only ever agree to do more Invader Zim work if it could explore the darker side more as he did when Nick canceled syndication. The Moofy was such a whimsical take on the dark foreshadowing of the way media and short lived celebrity would define the coming decades with social media.
And then there were the early Rick and Morty eps which contradict this where both the art is mediocre and generic, and even the characters are clear back to the future rip-offs but the plot and story and execution is so well done that it doesn't need a huge art budget (the Unity episode specifically).
I remember that Bugs Bunny became vanilla and humorless in the 60s as they began to do tired jokes and explained them in dialog, and no character was much different from any other. Somebody decided to make it palatable to 4-year-olds? I don't know. But never mind the perfect art; it was unflavored gelatin cartoon pablum.
I was mentored by several people who worked under him for Ren and Stimpy, and I’ve never met anyone who’s said a nice thing about working with him.
His blog is also one of the more toxic fixtures of the animation industry, leading to tons of people who look up to his style of critique in much the same way many engineers looked up to Linus’s early aggressive rants (but worse). He’s the proginator of the derogatory “Calarts bean mouth” description that so many wannabe artists throw around without thought, and led to so many death threats and harassing of brilliant artists.
That’s not even mentioning his rampant sexual abuse allegations and dating underage girls.
John K made some great content but man is he a blight on the industry as a whole, since Ren and Stimpy finished.
Meanwhile since you mentioned Genndy, who I’ve had the pleasure to work with directly, he’s an artistic genius and just great to work with. Genndy and Craig McCracken pushed an art style that defined multiple generations of cartoons, and Genndy pushed forms more than most people I know of.
A lot of the more fun stylization in CGI is thanks to Genndy pushing Sony hard. The first hotel Transylvania is mind blowing if you analyze the animation, and a huge part of that is Genndy drawing over frames to show the extreme silhouettes that he wants. Technology was then developed to allow artists to fit those forms repeatedly and reliably. That’s had knock on effects to other CG films, even ones from Disney and Pixar who analyzed how much he pushed forms. Not to mention he created three of the best shows: Dexters Lab, Samurai Jack and Primal. primal is a master class in story telling and cinematography.
Anyway long rant to say, man I dislike John K but Genndy is awesome.
It's not accurate to call it "AI generated animation". Face tracking is an input into the software that animates it. Those inputs feed into a bunch of manually configured math to move the thing around.
It's also not a new idea. The idea of having a human puppeteer a computer animation goes way back. What's new is streaming.
So this time I'm willing to believe that the small morphing clips and blurry NeRFs are going to grow into whole realistic 3d scenes.
Prediction: It's going to run into all sorts of copyright issues, wouldn't be perfect and make mistakes that a human would never do and end up being just another tool for Animators to use to create rough drafts that they can refine further.
computer tools already dominate the animation industry and have for a long time. Lots of that already is AI if we hadn't started to redefine what the word means. It's not going to cause unemployment, it's going to give animators more tools, and it will make animation more accessible.
There's a big gap between "Wow, this is really good for being made by a computer" and "Wow, this is consistently better than human experts".
At the same time, large businesses are very slow, so many companies will likely continue with traditional methods for decades while newer startups use the latest technology. It'll take decades before all business use AI pipelines. Animators should start their own studios to take advantage of the technology wave.
It's relatively straightforward to generate a procedural forest. Less so when the director wants "this branch" to curve more in a given direction.
Where I can see this having the greatest initial impact is in concept art and matte painting. And not in a good way for the artists.
When the engraving industry was automated demand went up.
People are creatures of habit and really don’t want to change our routines. Plus large businesses are extremely risk averse.
For the vast majority of animation that isn't just humanoids talking, or bodies moving, rigged cg animation is going to be better.
In fact, try comparing the control you get from rigged CG animation versus blindly groping through latent space for "just the right character with just the right animation"
I think in 5 years AI will dominate the meme generation "industry", but animation will still be dominated by good ol fashion rigged CGI.
The rigging and animation tools are what will be an order of magnitude better + faster, particularly facial and humanoid body stuff.
Animation isn't just about moving static images.
Typical neo-liberal ideology: "you can't do anything about capitalism".
Yes, you can. Humans controls AI, not the other way around (at least for now). Join your local communist party, organize to combat capitalism, discuss regulations to AI and corporations, worker ownership over the means of production, etc.
Another capitalist mouthpiece that thinks that AI, capitalism or human misery are natural events, like the rain or wind. There is nothing "inevitable" about this at all, capitalism doesn't grow in trees.
I am someone who has both worked in the animation industry (both as an artist and a developer), and currently work on what I would consider the bleeding edge of graphics engineering with ML (with large teams of course), I feel very confident in saying that people who make articles like this are just spewing technocratic bullshit.
There’s no understanding of the craft that goes into making content. There’s no understanding of the iterations and creative control that goes into it. There’s also no understanding of the business of animation.
Saying “you’ll lose your jobs” is downright stupid. Yes, AI tools will help streamline jobs. But that’s what they are: tools. Some jobs may be displaced if people don’t learn the new tools (see 2D animation being displaced by 3D) or if you were already just a worker doing menial work. However there’s so much artistry involved in every phase of production that you’re not going to see job reductions for a few decades.
I believe the opposite, there’s such a desire for a glut of content, that you’ll see more jobs. AI may help accelerate those tasks, but there’s already so much work that studios are constantly at capacity.
The other fact is that it ignores human nature. We still value stop motion animation or 2D even if CGI is better for mass production. There’s a human element that is not purely rational that we crave.
Lastly, the reason I think articles like this are bunk is because the people who write them assume studios do little research and development internally. We do so much. Attend any SIGGRAPH session and you’ll see so much mind blowing content.
We spend years trying to make it so artists don’t have to do work. I’ve literally automated my job and the job of other artists on other projects.
Does that mean my department or those artists get fired? No! It means they spend time working on making the content better instead of spending time on the stuff that was just automated. When they were spending time working on menial things like crowds, they can now spend that time finessing great performances.
We often stay on top of the cutting edge of research outside of our industry and make good use of it. It’s not like the industry, made up of countless brilliant engineers and artists, are blind to what’s going on.
Anyway, my advice to people: don’t listen to charlatans who sell technology as the answer to problem spaces they’ve never worked in. Yes, every now and then you get an outsider who shakes up the world, but it’s rare without some deeper understanding of the problem space.
The people saying Diffusion art generators and nerfs are coming for the animation jobs are quite clueless in my opinion, and are easily hyped by “shiny”.