While none of this is new information it does do a good job of surmising the current issues in the anime industry. Also found out because of this article that anime is now treated within the same scope as heavy industries like vehicle manufacturing.
LuminousVoxel2026-01-17 15:41
It definitely seems like the industry is pivoting more and more to haves and have-nots.
Series and movies that receive the correct attention are getting *really* good... meanwhile those that miss that bandwagon, even high profile examples, end up increasingly rough and unfinished...
VoidEmbracedWitch2026-01-17 16:05
If you're in the mood for some further reading about what causes the *broken* feel all too common in shots of modern anime, I'd recommend [The Layout Crisis](https://blog.sakugabooru.com/2022/08/05/the-layout-crisis-the-collapse-of-animes-traditional-immersion-and-the-attemps-to-build-it-anew/). Long story short, it's a lot of individual sources of pressure coming together to degrade the integrity of anime production pipelines over time.
BlackLilyOfTheMounts2026-01-17 19:33
This article was in 2022 and it' s crazy how prescient it is wtf.
VoidEmbracedWitch2026-01-17 19:39
The 2019 one about the [fragmentation of the anime industry](https://blog.sakugabooru.com/2019/08/31/the-fragmentation-of-anime-production-too-many-cooks-spoil-the-broth/) is also as relevant as ever when you look at staff credits of a show with rather rough layouts and see a dozen 2nd KAs, half a dozen directors of photography and another dozen animation directors for good measure.
Ordinal43NotFound2026-01-18 04:09
It's by KVin, legit one of the best sources for the ins-and-outs of the anime industry. Learned a lot about the production side of things from reading his blogposts.
detarameReddit2026-01-18 07:27
Sakugablog is like that. I wouldn't even call it prescient; just very perceptive. Like a Sherlock Holmes with a mental attic of everything anime production-related, if you will.
The anime industry has been subject to these forces for a while, and it is definitely getting worse...
NecroCannon2026-01-17 22:55
Holy shit, as an animator I’ve been doing layouts and not storyboards, that explains why I’m able to expand so much while the animation itself is suffering because I go right into drafting key frames
For those who didn’t read most of that and want an artist summarizing it. Basically imagine some of your favorite scenes that maybe has a bunch of different smaller actions happening during the sequence, like the final X scene in Silent Voice. That’s all drawn out and noted with what’s happening, this allows for storyboards to implement those, and animators to not be the only ones thinking about that. Without it you can get situations where like in One Piece, animators can basically do whatever in the scene as long as it fits and is approved. So that means really awesome scenes at its peak, and scenes that have good animation but something is just off, because the animator is the main one deciding what happens. A lack of layouts can also mean, things like a male character suddenly having boobs because it’s sent to someone that doesn’t understand the source material… *cough* One Punch Man *cough*
This means that anime production is becoming directionless, which when combined with a chaotic and crunch filled environment, also means that everything can be thrown off and the situation can hit a meltdown point. There’s no direction to point available resources to, leaving it up to the key animators, who could put a ton of effort into a scene but just there being a few of those can derail the production if they can’t handle the current situation. That’s how you end up with the sexy stuff having this good animation while fights seem lackluster, and that’s days, weeks, months, of time resources that could go to something that’s struggling more.
These massive studios that uses their manpower to pump out anime are leading the charge of this issue, because they genuinely don’t have time for layouts, don’t want to ruin their business model by investing money into it. Studios that are maybe medium sized with 1-2 money maker IPs are built on their attention to detail, more than likely still uses layouts because they only have maybe 1-2 shows in production at a given time. But they also can’t handle multiple failures so they have to really think about the work they take on and its execution.
But those few studios mean jack shit when there’s a ton of studios popping up, and their main focus is pumping out content that can sell enough. We’re essentially seeing the US VFX industry situation happening in Japan where the baseline is getting worse while the best is getting rare. If it keeps happening, the ones that win are the ones flooding the market, because there’s always a new low to just slightly avoid
Comprehensive_Dog6512026-01-18 03:31
The reason I follow that blog is for excellent content like that
UMP45isnotflat2026-01-19 01:36
great article, thx for the link
gc111172026-01-17 16:17
Pour one out for One Frame Man
dumpling-loverr2026-01-18 03:35
Not the first of its kind and won't be the last. The most recent high-profile/mainstream example before that was Boruto but that show has both animation and writing problems.
IKeepDoingItForFree2026-01-18 05:31
Yeah - I mean its also a key part of animes DNA from its inception as well with TezuPro. But you can easily go back decades and see the same issues, its clearly systematic to the industry thats only been magnified over time now and more people paying attention.
I know for example that Tatsunoko was infamous for enforcing very strict rules on both the number of drawings allowed for anime in both its storyboards and douga, and it also had a hard cap on the number of cels allowed per episode of a production - which resulted at times of the team having to drag cels across the capture pane to simulate movement as they quite literally not allowed to animate it as the cap for the episode was reached.
atropicalpenguin2026-01-17 17:34
Think that's been happening forever. We all remember the Macross and the Gundams, but how many others haven't withstood the test of time.
garfe2026-01-17 17:39
Hasn't that always been the issue with the industry though?
BlackLilyOfTheMounts2026-01-17 19:33
I think the craziest thing is that the industry is probably the most profittable ever and despite all of that, it has still the same issues.
Reemys2026-01-18 00:53
The number of viewers is increasing and translating into higher revenue, but no other parts of the equation are changing, alas.
IKeepDoingItForFree2026-01-18 05:37
You are pretty spot on. With how the production committee system works the studio unless they own the IP or also have bought into a spot on said committee would only get the payments from the committee to actually make the show (and any costs over the initial budget is out of pocket) and if they are lucky a really small residual (if any) - most of the revenue is getting split amongst the other committee members who bought in at the higher risk values or specific exclusive rights such as merchandise.
So having more revenue and more viewers at the end really doesn't translate directly to the studios bottom line and its why some studios go under despite being productive and having continual work lined up.
Reemys2026-01-18 10:15
This system is clearly the epitome of capitalism and benefits the corporations, but it dehumanises art production and does not nurture strong generations of writers, artists and animators in the industry. It sort-of min-maxes whatever exists already, with most people never getting a chance to work for a truly meaningful production.
Torque-A2026-01-17 19:27
wow it's almost like this is reflecting our current economy
Salty1452026-01-17 20:40
The rich get richer, and the poor get poorer.
Such is life.
stormblaz2026-01-17 23:48
My issue is we have so many mediocre anime, isekai and just slop that comes out yearly, and only 3-5 shows a season are actually worth people's time im afraid, unless you are a incredible anime Otaku, people struggle pushing through 12 episodes, so they usually watch what's worth.
I think studios need to filter and curate what manga or originals they are going to greenlit, if they want to stay healthy and profitable.
Iloveahrisears2026-01-18 11:35
I'd rather have the current state of anime than one where things pivot even harder into catering to the non-otakus. I'm already struggling to find meaningful stuff to watch, this wasn't the case 7 years ago.
Sw0rDz2026-01-18 00:36
One Punch Mans latest season comes to mind.
Reemys2026-01-18 00:52
Wouldn't this be the pattern for most of human activity by now? We could call it "talent" and we have haves and have-nots. With the amount of individual human units increasing exponentially the people pursuing various degrees and trades are also increasing - however, an individual human's productive life and remains the same. This leads to an extreme increase of supply, of mass of goods and services that demand cannot cover on a functional, physiological level. And market will not just solve this - humans cannot just stop functioning, they will have to keep being productive no matter their industry, with a few notable exceptions.
The market will kick out a % of workers in each industry and restructure itself by changing their occupation. But this seems to inevitably lead to people looking for the best cost/efficiency of their time spent, be it physical or mental goods, including fiction.
Doesn't it just seem that the world organically shifts towards junk-consumption and elite-consumption, while the in-between slowly withering as a concept?
Shadow_Ass2026-01-17 15:46
I already see how in a couple of years more and more studios will use AI to replace some animators because people are getting sick or not wanting to work in an industry which makes a bunch of money but doesn't treat the people who make it possible right. Greed is fucking us all up
TastyOreoFriend2026-01-17 16:18
It's sad unfortunately because generative AI is already garnering a terrible reputation for issues with theft. Using it is treating the symptoms and not the root cause. Once the anime industry figures out their model with AI it's going to cause a rush to start using it to remain competitive.
It's going to be a race to the bottom if you still appreciate art made by humans.
StickiStickman2026-01-17 19:22
There's no one who thinks it's ""theft"" to learn from publicly available pictures outside of a tiny Reddit minority.
There's been a dozen high profile court cases already with the same outcome.
Also, SO MUCH of the work is already insanely repetitive and grueling manual labor. It absolutely should be automated if possible.
_trouble_every_day_2026-01-17 19:34
Fuck off dilettante
BlackLilyOfTheMounts2026-01-17 19:34
It' s AI slop. Repetive work can be automated, but the people selling you AI right now wants to straight up replace you if they can.
TastyOreoFriend2026-01-17 19:54
This is the thing that I wish more people who champion AI would understand. The current AI models are not for your benefit—shareholders who don't want overhead from human resources are the winners there.
It'll be no different with anime all while making it more difficult for animators to push for better wages, benefits and working conditions. You can pretty much apply the same logic to any other industry that AI will affect as well.
StickiStickman2026-01-17 20:15
Not everything made with AI is slop and not everything made by hand is good. Just look at the slop that is One Punch Man.
> but the people selling you AI right now wants to straight up replace you if they can.
Since I'm a programmer, yea. But I also use it daily since it's a super useful too. You're just against any and all automation and industrialization? Because automating tedious soul crushing work is a good thing.
potatojayy2026-01-20 00:26
Uh One Punch Man dipped in quality due to how its studio cut corners after season 1 by considerably lowering the budget and severely tightening its schedule putting animators under duress. So it's not the animators' fault... it's the higher ups creating a stressful environment trickling down to the subpar final product - an issue that could've been avoided had they not cut corners to begin with. Maybe the soul crushing part of the job is the industry as a whole, not the actual work itself.
Using One Punch Man as an example did not support your argument at all.
StickiStickman2026-01-21 01:33
So it's not slop even though it was literally pushed out as fast as possible for no care of quality? Even though that's pretty much the exact definition?
Just admit you don't care and just want to have a crusade.
potatojayy2026-01-21 09:06
Such an ironic last statement seeing as how you're hard on a crusade already lol.
Identifying OPM's slop wasn't the crux of my argument. What i'm saying said human made anime could've had much better quality under better conditions, hence no need for AI. Using OPM as the example for human slop wasn't a fair example given what went on behind the scenes.
TastyOreoFriend2026-01-17 19:45
> There's no one who thinks it's ""theft"" to learn from publicly available pictures outside of a tiny Reddit minority.
Funny because the amount of people I see try to play this angle of "minority" when it comes to this issue are people who are on social media like reddit.
>There's been a dozen high profile court cases already with the same outcome.
Yes which is losing with massive settlements. Anthropic alone just settled for 1.5 billion last year. They've been losing left and right on this 🤣🤣🤣
>Also, SO MUCH of the work is already insanely repetitive and grueling manual labor. It absolutely should be automated if possible.
Whether or not it's repetitive/grueling is a another issue entirely. They're still jobs and the people who do them deserve to paid a decent wage rather than replaced by a bot.
Anzereke2026-01-17 18:26
Eh, I think the more likely outcome is that gen AI becomes like 3d, where it becomes a sign of a production cheaping out and is disliked on principle.
Some tools behind the scenes may help, but honestly I'd expect the gains to be coming from places (meaning other sorts of 'AI' tools) other than genAI. As art tools go, it's pretty awful in all the ways that an anime production needs it not to be.
StickiStickman2026-01-17 19:17
I can see it working really well for helping with in-betweens of key frames.
Umr_at_Tawil2026-01-17 20:29
This might be controversial but I think having AI do in-betweens would be one of the best use for it.
in-between already being paid slave wage right now, so bad that much of it is being outsourced to sweatshops in South-East Asia all the time (you can see many SEA region name like Vietnamese name in pretty much all anime credit), and sometimes those outsource company [covertly outsource them to North Korea too](https://www.38north.org/2024/04/what-we-learned-inside-a-north-korean-internet-server-how-well-do-you-know-your-partners/). It's also very grueling time-consuming works.
having AI do in-between would save all this time, and might allow key animator to experiment more, coming up with better shots without the time pressure of sending it out ASAP so in-between could be done in time for the airing.
while in-between was used as training for animator before, we could have animation school teach people straight into key animator level instead.
[Just look at how much AI brushed up the animation of episode 6 of OPM S3](https://np.reddit.com/r/OPMFolk/comments/1p38qv7/episode_6_ai_vs_official/)
r4wrFox2026-01-18 11:15
I like how you acknowledge a problem, the underpaying/outsourcing in an important role for not just the quality of an anime but the training of future generations of animators, and see the solution as simply automating that position away.
You can't simply school someone into more senior positions, because industry experience cannot be taught. That is why its called Experience. These junior animators will enter key animation positions as the next most junior position, but their key animation will not be up to the same quality as those who had experienced the industry beforehand.
This causes one of two things to happen: either those higher up the chain have to cover for the shoddy work, by heavily editing or near completely redrawing the animation of the key animator, thus creating extra redundant work on higher ranking staff, or the key frames would get sent to the AI uncritically, which would interpolate whatever was fed into it and lead to a noticeably lower quality product independent of the quality of the AI (which everyone knows will be shit, but we can pretend for sake of ai bro hypotheticals that it would be the same as a human).
In both of these scenarios, you're either making the working experience more difficult/stressful for everyone involved (no one likes having their work redone, and no one likes redoing work they delegated to another person), or you're lowering the quality of the end product dramatically.
And all these negative effects, only to realistically not save any money. Because AI is actually really expensive to run at scale. You've either got the costs of literally erecting a data center, which is incredibly expensive to build/maintain, or you're paying for what could only be an incredibly expensive usage plan when you account for the sheer amount of in-betweens you'd be asking ai to produce. The fact that all of these AI companies are run by the biggest players in the market, yet none of them have figured out how to make money with their tech should speak to the economic impracticality of AI as it exists.
There's a reason it's a v common joke in tech spaces to claim AI stands for "Actually Indians." Because it is infinitely cheaper to outsource shit out to places in low cost of living areas, like India or Southeast Asia, than it is to actually use artificial intelligence.
AI is yet again a solution looking for a problem, in a place no one fucking wants it.
Umr_at_Tawil2026-01-18 20:07
>You can't simply school someone into more senior positions, because industry experience cannot be taught. That is why its called Experience. These junior animators will enter key animation positions as the next most junior position, but their key animation will not be up to the same quality as those who had experienced the industry beforehand.
You can though, it's called extended, on-the-job training, how do you think job like surgeon train their junior? having the junior watching the senior work then having junior doing the job itself under supervision of the senior. this is what already in place in most studio in Japan because the low wage and widespread outsourcing anyway.
>This causes one of two things to happen: either those higher up the chain have to cover for the shoddy work, by heavily editing or near completely redrawing the animation of the key animator, thus creating extra redundant work on higher ranking staff, or the key frames would get sent to the AI uncritically, which would interpolate whatever was fed into it and lead to a noticeably lower quality product independent of the quality of the AI (which everyone knows will be shit, but we can pretend for sake of ai bro hypotheticals that it would be the same as a human).
This is very a very ignorant and arrogant assumption of AI capability, what make you think it's always gonna be worse? look at what I linked, look at what AI was able to do with what it was given from episode 6 of OPM S3, you can find many other examples on youtube and it has already changed many people opinion on AI, and that was with a general, publicly available AI too, specialized AI gonna do even better, and it only going to get better with time.
what you are saying no different from insisting that the sky is green while it's blue. saying as if this is a matter of course without it being so.
>And all these negative effects, only to realistically not save any money. Because AI is actually really expensive to run at scale. You've either got the costs of literally erecting a data center, which is incredibly expensive to build/maintain, or you're paying for what could only be an incredibly expensive usage plan when you account for the sheer amount of in-betweens you'd be asking ai to produce. The fact that all of these AI companies are run by the biggest players in the market, yet none of them have figured out how to make money with their tech should speak to the economic impracticality of AI as it exists.
Economy of scale would make it cheaper, not more expensive. Much of the cost of AI right now is in infrastructure buildout, research and training, not inference, i.e using the AI itself, there will be a point where this slow down and cost would also go down as well. This is already partially happen in China, where they have much lower cost, from personnel to power to the infrastructure itself, with Deepseek already reporting positive cashflow.
>There's a reason it's a v common joke in tech spaces to claim AI stands for "Actually Indians." Because it is infinitely cheaper to outsource shit out to places in low cost of living areas, like India or Southeast Asia, than it is to actually use artificial intelligence.
That joke is actually apply to position outside of software development, where AI is not good enough to replace human yet (like accounting, customer support and content moderation), and sometimes, also mean Indian software developer who is much cheaper to hire but have similar quality and productivity thanks to the help of AI.
why I know this? because I am a senior software engineer with 9 years of experience, AI has helped me greatly at my job, it made my productivity went through the roof along with all of my co-workers, it has allowed my company to take on more projects, even with tech we're unfamiliar with. for example, none of us ever written something as complex as a compiler, and before AI we would never dare to take on such a project, but with the help of AI (Claude Code specifically) we were able to successfully develop and ship this project. This has made my company a lot more money in the last 2 years and I got a big raise out thanks to that. With supervision and feedback, AI actually helped improve the quality and maintainability of our code.
>AI is yet again a solution looking for a problem, in a place no one fucking wants it.
It is already the solution to many of our problems, like many of mine, and Terence Tao, one of the greatest mathematician alive, has praised AI for how helpful it is with math. AI has already proven itself regardless of what alternative reality you're dreaming up. Adoption of AI into anime production is inevitable, just like the adoption of 3D for reference and into the final cut of the anime itself.
I suggest you should look at this with a more balanced perspective, outside of the echo chamber of negativity that's /r/Technology.
r4wrFox2026-01-19 00:33
> no one on our team had ever made something as complex as a compiler
... bro they teach undergrad students to do this one. The fact you needed AI to even try kinda, says more than I really need to add to this conversation.
Umr_at_Tawil2026-01-19 01:07
Those are toy compilers for simple language (I did it too), not a recursive-descent compiler for a complex proprietary language with C++-like syntax written from scratch, with many of its own special quirks and rule. There is a reason why they need to commission this custom compiler from us.
Go ask anyone who's working professionally in software development, ask them if they can write a commercial-use compiler from scratch, they will rank it as one of the more difficult task out there.
It's like, you are saying just because you're taught basic algebra and calculus in high school, you can do complex, frontier math like an University math professor.
Anzereke2026-01-18 20:14
I'll believe it when someone actually gets it working. Right now we have a lot of people promising this sort of stuff is, say it with me now, right around the corner.
Avatar_exADV2026-01-17 20:22
I mean... this isn't a bad thing, strictly speaking. Drawing in-betweens is scut work. It's not a creative endeavor, it's repetitive drudgery that has to be done by a person who could otherwise be doing creative things because that's the way you develop the necessary technical skill. If you can draw just the key frames and then say "okay, compy-chan, draw my in-betweens", that lets a lot of additional people produce high-quality animation (or at least, as high-quality as their key frames, but of course that's an issue today too...)
It doesn't mean you can produce an ANIME cheaply because you still have to do all the audio work, and of course computers aren't quite there for generating voice acting yet either. But saying "it's terrible that people won't be drawing in-between frames by hand!" is right up there with "how can we have good clothing if people aren't combing out the cotton by hand?" Gin that stuff, man!
PointmanW2026-01-17 20:47
I don't think it's Greed personally.
when you think about it, anime is fking hard to monetize for how expensive it is to make.
with video game, you just sell copy and that's money to you, very straightforward, if it sell well then you're rich.
with anime, very few people is willing to pay even a small amount of money for a single anime that they might not like, that leave streaming as the only option, and payout for it is not good enough to cover the production cost.
that why anime have the production committee model in the first place, where different companies invest into an IP to make an anime, hoping that the increased publicity would make them more money on merchadise, music and and other multimedia stuff, but the anime by itself is almost never profitable.
wilkened0052026-01-18 01:40
The anime industry is already profitable on its own show. The problem is that the industry is constantly overheating with no time to rest as it struggles to keep up with the growing global demand for anime.
ChiggaOG2026-01-17 23:12
The middleman get significant amount of money.
idolMechaFandesu2026-01-17 15:48
[Business as usual](https://tenor.com/ldwaFZkbkIa.gif)
Sweet-Message11532026-01-17 15:57
nothing exemplified "corporate greed kills quality" more than OPM s3
Jonathank922026-01-17 16:11
they could've kept season 3 honestly. Wait another 5 years until a better studio is available i don't care.
dumpling-loverr2026-01-18 03:41
What the fuck is this recency bias. OPM S3 is the latest high profile victim when this budgeting and overworking issue has been an issue with the industry for decades now.
7DS, Blue Lock, Boruto, Berserk (2016), Promised Neverland s2, etc.
Sweet-Message11532026-01-18 03:59
7DS had some good moments
Blue Lock had 5mins of actual quality animation
BERSERK had a killer OST
Promised Neverland s2 never happend
Boruto is overhated
OPM s3 literally failed in every department from animation, storyboard, music, editing & voice acting
every series you've mentioned have either ended or have improved...
OPM s3 production team are not only defending that sh!t but also announced a continuation
dumpling-loverr2026-01-18 05:30
Revisionist take. They were all hated either way and deservedly trashed during their time. This has been a problem of the Japanese animation industry for 3 decades now.
There's a reason why Chinese studios are grabbing talented animators and producing quality anime games + shows while Japan stagnated and is MASSIVELY left behind.
notreal0882026-01-17 16:20
Pay your people hire more people and stop being greedy. It’s not the studios that are messing things up. It’s the committees who green light the budget for anime and decide that Pennie off the dollar should be enough for the studio.
This is why studios like whitefox go under. They can only make real money by selling merch. The production of the anime if not a loss it’s pretty damn close.
If the pie were better distributed the studios could possibly hire more people and pay their works more
SakuraNeko72026-01-18 06:05
I agree with paying people better but hiring more people isn't just an easy thing and has its limits. You can't hire 3 people to change a lightbulb faster for example.
Maqoba2026-01-17 16:30
So you're telling me anime studios are businesses and will employ the same tactics as other businesses. I'm shocked.
youravgindian2026-01-18 07:12
This article sounds more like 'yes, capitalism exists'. Not defending it any shape or form, but real reforms need to happen instead of pumping out 678th article about the 'issue'.
DAiDAiDa2026-01-17 16:49
There is even a active crashout from one of the praised animator in indistury. Weilin Zhang who is debuted at 16 year old in boruto anime with naruto with sasuke hand to hand combat with alien. Low payments unsustainable job security and indistury behaviour. And wanting to change his career.
thereturnofghettopat2026-01-18 05:32
i find it hard to disagree with most of what he expressed in that blog post, but wtf was with the chainsaw man stray 😭
ShielderKnight2026-01-17 17:43
I wish instead of 80 anime a season we just go back to 30 or 35 max and increase the work and quality on the 30/35
UltimateKaiser2026-01-17 21:06
This won’t happen because of the move from long running shows to seasonal. There’s a ‘quota’ now
SubmissiveDinosaur2026-01-17 21:59
Also, if there's a lot of animes its because there are more studios in the game now, each competing for a spot in the best TV airing schedules (or trying to break trough stream platforms)
IKeepDoingItForFree2026-01-18 05:39
If you really want to get old head - Summer 1993 had exactly 1 new show air, everything else was continuation.
UltimateKaiser2026-01-18 17:04
That’s crazy fr
Magnus-Artifex2026-01-18 12:41
As much as I wish it was different, it doesn’t quite work like that
ElliotAlderson20242026-01-17 17:52
Perhaps stop pumping out endless fantasy/isekai garbage every season.
Jusenkyo_52026-01-17 17:54
Too much garbage is coming out that barely turns a profit if at all. Instead of 15 villainess anime or another anime where a character has an overpowered cheat skill we could utilize those same people to work on shows that people actually like and watch.
Follow the money. The production committee system needs to be destroyed or overhauled, the anime industry can't continue to exist forever being 80% "read the manga" commercial projects.
Whendfield1232026-01-17 18:02
Hey dont put villainess animes into this. I want more of it
Jusenkyo_52026-01-17 18:06
😂😂 In all fairness I haven't seen many, it just seems like the genre exploded after My Life as a Villainess came out.
alotmorealots2026-01-18 04:14
You're missing out, the genre is still going through the phase where the best of the material is being adapted, albeit with low production budgets because of the target audience size (and demographics). There are a lot of reused tropes that form the genre's structure, but within that are exceptionally dynamic and diverse female leads. They're also frequently very funny!
Jusenkyo_52026-01-18 04:28
Which would you recommend? I will watch literally anything if it's good or unique.
DrStein10102026-01-18 04:25
I love Villainess slop. I don't even care that most of it is shit.
Abysswatcherbel2026-01-17 19:23
>same people to work on shows that people actually like and watch.
People that share your opinion are not going to enjoy the choice of shows that "people actually like and watch"
Jusenkyo_52026-01-17 20:12
I knew that someone would call me out on that one lol.
I'm not advocating for the Sword Art Online's or the Reincarnated as a Slime's of the world to not be made, but I think it's near undeniable that there are 10-20 shows every season that get like 0.5% viewership and end up rated as a 6 on MyAnimeList. I'm not the arbiter of anime taste and I don't think the industry should cater to me.
Whendfield1232026-01-17 18:04
Maybe the aime industry should have a union. Its also an important soft power for japan, so the government should take care to ensure the industry to decline.
Kadmos12026-01-17 18:56
It is almost like a darned if you do and darned if you don't. That is, if an anime studio works on more titles that might be good for business in terms of staying afloat. However, the cost of over-worked and under-paid animators is concerning. Alternatively, less output might be bad for business in staying afloat and is a double-edged sword for animators. What I mean is that while less anime might make their lives less stressful, it could also mean less money for them to earn.
MastodonParking90802026-01-17 19:11
Kinda funny how the period (2007-2014) where international consumption was it's lowest was also the time when classics like Code Geass, Madoka, Gurren Lagann, Steins;Gate etc were being pumped out every season...
Stofenthe1st2026-01-17 19:58
Well it’s not like they’re still not releasing new hits. Frieren, Delicious in Dungeon, Chainsaw Man, Jujustu Kaisen, One Punch Man(season 1), Mob Psycho 100, etc. You get the picture.
The talent is still there but it’s being squeezed dry because of the workload of modern anime. If there’s not something done then there might not be a workforce to actually make anime.
UMP45isnotflat2026-01-19 01:42
the noteworthy part here is that all of those are original anime, except for Steins;Gate. Now it feels like most things are just adaptions.
MastodonParking90802026-01-19 03:40
It's not just that, the writing in many of the shows OP listed like JJK is agreed upon by fans themselves to be non-existent. Well, beyond just the lack of new aesthetics or ambitious worldbuilding, perhaps it's lack of mythical quality to it all.
UMP45isnotflat2026-01-19 15:58
you didnt even mention JJK in your original comment though?
The_Spicy_brown2026-01-17 21:38
Interesting article. Nothing new, but well condensed.
almisami2026-01-17 23:10
As someone who grew up with the GORGEOUS hand drawn machines in OVSa from the 90s, I believe the industry would benefit from stepping back in scope and focusing on quality.
You'd sell more merch if your product was quality. I hardly remember any characters from the last 5 years of anime except Frieren.
ODST_Parker2026-01-18 08:25
As someone who only really got into anime in the last five years, I remember dozens of great shows, and what feels like a hundred great characters off the top of my head.
That said, I agree that a focus on quality over quantity is important, as long as it means there's less stress on the production timeline.
almisami2026-01-18 08:38
I'm open to recommendations.
sonosabi2026-01-18 10:29
The same thing that happened in the game industry is now happening here. Because audiences have seen so many beautifully animated shows, anything with weaker animation is automatically seen as low quality. This has pushed studios into a kind of “**Sakuga chicken race.**” The result is more anime that looks great but isn’t actually interesting, while smaller studios keep collapsing. Meanwhile, major publishers are also fighting over manga to adapt, IP ownership, and talented animators.
Commentaires
Series and movies that receive the correct attention are getting *really* good... meanwhile those that miss that bandwagon, even high profile examples, end up increasingly rough and unfinished...
The anime industry has been subject to these forces for a while, and it is definitely getting worse...
For those who didn’t read most of that and want an artist summarizing it. Basically imagine some of your favorite scenes that maybe has a bunch of different smaller actions happening during the sequence, like the final X scene in Silent Voice. That’s all drawn out and noted with what’s happening, this allows for storyboards to implement those, and animators to not be the only ones thinking about that. Without it you can get situations where like in One Piece, animators can basically do whatever in the scene as long as it fits and is approved. So that means really awesome scenes at its peak, and scenes that have good animation but something is just off, because the animator is the main one deciding what happens. A lack of layouts can also mean, things like a male character suddenly having boobs because it’s sent to someone that doesn’t understand the source material… *cough* One Punch Man *cough*
This means that anime production is becoming directionless, which when combined with a chaotic and crunch filled environment, also means that everything can be thrown off and the situation can hit a meltdown point. There’s no direction to point available resources to, leaving it up to the key animators, who could put a ton of effort into a scene but just there being a few of those can derail the production if they can’t handle the current situation. That’s how you end up with the sexy stuff having this good animation while fights seem lackluster, and that’s days, weeks, months, of time resources that could go to something that’s struggling more.
These massive studios that uses their manpower to pump out anime are leading the charge of this issue, because they genuinely don’t have time for layouts, don’t want to ruin their business model by investing money into it. Studios that are maybe medium sized with 1-2 money maker IPs are built on their attention to detail, more than likely still uses layouts because they only have maybe 1-2 shows in production at a given time. But they also can’t handle multiple failures so they have to really think about the work they take on and its execution.
But those few studios mean jack shit when there’s a ton of studios popping up, and their main focus is pumping out content that can sell enough. We’re essentially seeing the US VFX industry situation happening in Japan where the baseline is getting worse while the best is getting rare. If it keeps happening, the ones that win are the ones flooding the market, because there’s always a new low to just slightly avoid
I know for example that Tatsunoko was infamous for enforcing very strict rules on both the number of drawings allowed for anime in both its storyboards and douga, and it also had a hard cap on the number of cels allowed per episode of a production - which resulted at times of the team having to drag cels across the capture pane to simulate movement as they quite literally not allowed to animate it as the cap for the episode was reached.
So having more revenue and more viewers at the end really doesn't translate directly to the studios bottom line and its why some studios go under despite being productive and having continual work lined up.
Such is life.
I think studios need to filter and curate what manga or originals they are going to greenlit, if they want to stay healthy and profitable.
The market will kick out a % of workers in each industry and restructure itself by changing their occupation. But this seems to inevitably lead to people looking for the best cost/efficiency of their time spent, be it physical or mental goods, including fiction.
Doesn't it just seem that the world organically shifts towards junk-consumption and elite-consumption, while the in-between slowly withering as a concept?
It's going to be a race to the bottom if you still appreciate art made by humans.
There's been a dozen high profile court cases already with the same outcome.
Also, SO MUCH of the work is already insanely repetitive and grueling manual labor. It absolutely should be automated if possible.
It'll be no different with anime all while making it more difficult for animators to push for better wages, benefits and working conditions. You can pretty much apply the same logic to any other industry that AI will affect as well.
> but the people selling you AI right now wants to straight up replace you if they can.
Since I'm a programmer, yea. But I also use it daily since it's a super useful too. You're just against any and all automation and industrialization? Because automating tedious soul crushing work is a good thing.
Using One Punch Man as an example did not support your argument at all.
Just admit you don't care and just want to have a crusade.
Identifying OPM's slop wasn't the crux of my argument. What i'm saying said human made anime could've had much better quality under better conditions, hence no need for AI. Using OPM as the example for human slop wasn't a fair example given what went on behind the scenes.
Funny because the amount of people I see try to play this angle of "minority" when it comes to this issue are people who are on social media like reddit.
>There's been a dozen high profile court cases already with the same outcome.
Yes which is losing with massive settlements. Anthropic alone just settled for 1.5 billion last year. They've been losing left and right on this 🤣🤣🤣
>Also, SO MUCH of the work is already insanely repetitive and grueling manual labor. It absolutely should be automated if possible.
Whether or not it's repetitive/grueling is a another issue entirely. They're still jobs and the people who do them deserve to paid a decent wage rather than replaced by a bot.
Some tools behind the scenes may help, but honestly I'd expect the gains to be coming from places (meaning other sorts of 'AI' tools) other than genAI. As art tools go, it's pretty awful in all the ways that an anime production needs it not to be.
in-between already being paid slave wage right now, so bad that much of it is being outsourced to sweatshops in South-East Asia all the time (you can see many SEA region name like Vietnamese name in pretty much all anime credit), and sometimes those outsource company [covertly outsource them to North Korea too](https://www.38north.org/2024/04/what-we-learned-inside-a-north-korean-internet-server-how-well-do-you-know-your-partners/). It's also very grueling time-consuming works.
having AI do in-between would save all this time, and might allow key animator to experiment more, coming up with better shots without the time pressure of sending it out ASAP so in-between could be done in time for the airing.
while in-between was used as training for animator before, we could have animation school teach people straight into key animator level instead.
[Just look at how much AI brushed up the animation of episode 6 of OPM S3](https://np.reddit.com/r/OPMFolk/comments/1p38qv7/episode_6_ai_vs_official/)
You can't simply school someone into more senior positions, because industry experience cannot be taught. That is why its called Experience. These junior animators will enter key animation positions as the next most junior position, but their key animation will not be up to the same quality as those who had experienced the industry beforehand.
This causes one of two things to happen: either those higher up the chain have to cover for the shoddy work, by heavily editing or near completely redrawing the animation of the key animator, thus creating extra redundant work on higher ranking staff, or the key frames would get sent to the AI uncritically, which would interpolate whatever was fed into it and lead to a noticeably lower quality product independent of the quality of the AI (which everyone knows will be shit, but we can pretend for sake of ai bro hypotheticals that it would be the same as a human).
In both of these scenarios, you're either making the working experience more difficult/stressful for everyone involved (no one likes having their work redone, and no one likes redoing work they delegated to another person), or you're lowering the quality of the end product dramatically.
And all these negative effects, only to realistically not save any money. Because AI is actually really expensive to run at scale. You've either got the costs of literally erecting a data center, which is incredibly expensive to build/maintain, or you're paying for what could only be an incredibly expensive usage plan when you account for the sheer amount of in-betweens you'd be asking ai to produce. The fact that all of these AI companies are run by the biggest players in the market, yet none of them have figured out how to make money with their tech should speak to the economic impracticality of AI as it exists.
There's a reason it's a v common joke in tech spaces to claim AI stands for "Actually Indians." Because it is infinitely cheaper to outsource shit out to places in low cost of living areas, like India or Southeast Asia, than it is to actually use artificial intelligence.
AI is yet again a solution looking for a problem, in a place no one fucking wants it.
You can though, it's called extended, on-the-job training, how do you think job like surgeon train their junior? having the junior watching the senior work then having junior doing the job itself under supervision of the senior. this is what already in place in most studio in Japan because the low wage and widespread outsourcing anyway.
>This causes one of two things to happen: either those higher up the chain have to cover for the shoddy work, by heavily editing or near completely redrawing the animation of the key animator, thus creating extra redundant work on higher ranking staff, or the key frames would get sent to the AI uncritically, which would interpolate whatever was fed into it and lead to a noticeably lower quality product independent of the quality of the AI (which everyone knows will be shit, but we can pretend for sake of ai bro hypotheticals that it would be the same as a human).
This is very a very ignorant and arrogant assumption of AI capability, what make you think it's always gonna be worse? look at what I linked, look at what AI was able to do with what it was given from episode 6 of OPM S3, you can find many other examples on youtube and it has already changed many people opinion on AI, and that was with a general, publicly available AI too, specialized AI gonna do even better, and it only going to get better with time.
what you are saying no different from insisting that the sky is green while it's blue. saying as if this is a matter of course without it being so.
>And all these negative effects, only to realistically not save any money. Because AI is actually really expensive to run at scale. You've either got the costs of literally erecting a data center, which is incredibly expensive to build/maintain, or you're paying for what could only be an incredibly expensive usage plan when you account for the sheer amount of in-betweens you'd be asking ai to produce. The fact that all of these AI companies are run by the biggest players in the market, yet none of them have figured out how to make money with their tech should speak to the economic impracticality of AI as it exists.
Economy of scale would make it cheaper, not more expensive. Much of the cost of AI right now is in infrastructure buildout, research and training, not inference, i.e using the AI itself, there will be a point where this slow down and cost would also go down as well. This is already partially happen in China, where they have much lower cost, from personnel to power to the infrastructure itself, with Deepseek already reporting positive cashflow.
>There's a reason it's a v common joke in tech spaces to claim AI stands for "Actually Indians." Because it is infinitely cheaper to outsource shit out to places in low cost of living areas, like India or Southeast Asia, than it is to actually use artificial intelligence.
That joke is actually apply to position outside of software development, where AI is not good enough to replace human yet (like accounting, customer support and content moderation), and sometimes, also mean Indian software developer who is much cheaper to hire but have similar quality and productivity thanks to the help of AI.
why I know this? because I am a senior software engineer with 9 years of experience, AI has helped me greatly at my job, it made my productivity went through the roof along with all of my co-workers, it has allowed my company to take on more projects, even with tech we're unfamiliar with. for example, none of us ever written something as complex as a compiler, and before AI we would never dare to take on such a project, but with the help of AI (Claude Code specifically) we were able to successfully develop and ship this project. This has made my company a lot more money in the last 2 years and I got a big raise out thanks to that. With supervision and feedback, AI actually helped improve the quality and maintainability of our code.
>AI is yet again a solution looking for a problem, in a place no one fucking wants it.
It is already the solution to many of our problems, like many of mine, and Terence Tao, one of the greatest mathematician alive, has praised AI for how helpful it is with math. AI has already proven itself regardless of what alternative reality you're dreaming up. Adoption of AI into anime production is inevitable, just like the adoption of 3D for reference and into the final cut of the anime itself.
I suggest you should look at this with a more balanced perspective, outside of the echo chamber of negativity that's /r/Technology.
... bro they teach undergrad students to do this one. The fact you needed AI to even try kinda, says more than I really need to add to this conversation.
Go ask anyone who's working professionally in software development, ask them if they can write a commercial-use compiler from scratch, they will rank it as one of the more difficult task out there.
It's like, you are saying just because you're taught basic algebra and calculus in high school, you can do complex, frontier math like an University math professor.
It doesn't mean you can produce an ANIME cheaply because you still have to do all the audio work, and of course computers aren't quite there for generating voice acting yet either. But saying "it's terrible that people won't be drawing in-between frames by hand!" is right up there with "how can we have good clothing if people aren't combing out the cotton by hand?" Gin that stuff, man!
when you think about it, anime is fking hard to monetize for how expensive it is to make.
with video game, you just sell copy and that's money to you, very straightforward, if it sell well then you're rich.
with anime, very few people is willing to pay even a small amount of money for a single anime that they might not like, that leave streaming as the only option, and payout for it is not good enough to cover the production cost.
that why anime have the production committee model in the first place, where different companies invest into an IP to make an anime, hoping that the increased publicity would make them more money on merchadise, music and and other multimedia stuff, but the anime by itself is almost never profitable.
7DS, Blue Lock, Boruto, Berserk (2016), Promised Neverland s2, etc.
Blue Lock had 5mins of actual quality animation
BERSERK had a killer OST
Promised Neverland s2 never happend
Boruto is overhated
OPM s3 literally failed in every department from animation, storyboard, music, editing & voice acting
every series you've mentioned have either ended or have improved...
OPM s3 production team are not only defending that sh!t but also announced a continuation
There's a reason why Chinese studios are grabbing talented animators and producing quality anime games + shows while Japan stagnated and is MASSIVELY left behind.
This is why studios like whitefox go under. They can only make real money by selling merch. The production of the anime if not a loss it’s pretty damn close.
If the pie were better distributed the studios could possibly hire more people and pay their works more
Follow the money. The production committee system needs to be destroyed or overhauled, the anime industry can't continue to exist forever being 80% "read the manga" commercial projects.
People that share your opinion are not going to enjoy the choice of shows that "people actually like and watch"
I'm not advocating for the Sword Art Online's or the Reincarnated as a Slime's of the world to not be made, but I think it's near undeniable that there are 10-20 shows every season that get like 0.5% viewership and end up rated as a 6 on MyAnimeList. I'm not the arbiter of anime taste and I don't think the industry should cater to me.
It is almost like a darned if you do and darned if you don't. That is, if an anime studio works on more titles that might be good for business in terms of staying afloat. However, the cost of over-worked and under-paid animators is concerning. Alternatively, less output might be bad for business in staying afloat and is a double-edged sword for animators. What I mean is that while less anime might make their lives less stressful, it could also mean less money for them to earn.
The talent is still there but it’s being squeezed dry because of the workload of modern anime. If there’s not something done then there might not be a workforce to actually make anime.
You'd sell more merch if your product was quality. I hardly remember any characters from the last 5 years of anime except Frieren.
That said, I agree that a focus on quality over quantity is important, as long as it means there's less stress on the production timeline.