No need. Soon there will be so much AI generated "art" that human-made art will largely be pushed out of popular consciousness.
People won't stop creating just because there's a cheaper alternative.
The analogy with manual labor automation simply doesn't hold, first because people actually like creating stuff, but perhaps most importantly because there is no supply problem to solve.
Taking the analogy with automation in the car industry, automation drastically lowered prices, making each individual copy of a product accessible to anyone.
It costs virtually 0$ to copy a movie, a picture, a song and display it on every device on the planet.
The cost of creating a audiovisual product is amortized across all users, dividing the costs by the number of users instead of multiplying it.
Sure, one day we may have models capable of generating full 3d animated movies from just a prompt, paying for just the electricity used to run the GPUs, but how much cheaper can you get compared to 9$/month to compete with a traditional streaming subscription with human-made content?
Assuming the world even comes to the point where AI streaming services contain virtually infinite amounts of meaningless autogenerated content, and assuming people actually like that content (which may be the case for the chunk of people already watching statistics-driven garbage on Netflix), nothing will stop creatives fired from previously-human studios from creating their own studios, producing hand-crafted, human works at the same price as always (9$/month).
An analogy can be made with handcrafted cars like Lamborghinis may cost millions, with the difference that with audiovisual content, a work can be shared by (and the price amortized by) billions.
Given the alternative of cheaper, semiautomated 2d animation, anime studios prefer handcrafting every single frame, instead, because some people really like drawing, and a large amount of people also likes watching handcrafted, high-quality animation with wonderful stories and visuals.
They care about the outcome not the process. They care that it looks how they want it to look, that it conveys what they want it to convey. This is artistic pragmatism at its finest. They will composite hand drawn graphics/animation, 2D computer graphics/animation, and 3D computer graphics/animation… it’s the final results that matter. I absolutely expect that once the tooling gets better at frame-to-frame consistency, once it is good enough AI/ML based animation tools will be widely adopted by the “anime industry” and it will be praised because it will do things like let small manga creators produce short OVA or movies or YouTube style content directly whenever they want without having to spend ages working with an animation studio and the fans will love it because they love the content and for the most part that love is not dependent on how the content was made.
Not that I have anything against cel shaded 3d, it looks very nice when done properly, but I still see no major shift from hand-drawn animation in the anime industry.
