We are about six years into transformer models. By now we can get transformers to write coherent short stories, and you can get to novel lengths with very careful iterative prompting (e.g. let the AI generate an outline, then chapter summaries, consistency notes, world building, then generate the chapters). But to get anything approaching a good story you still need a lot of manual intervention at all steps of the process. LLMs go off the rail, get pacing completely wrong and demonstrate gaping holes in their understanding of the real world. Progress on new models is mostly focused in other directions, with better storytelling a byproduct. I doubt we get to "best screenplay" level of writing in five years.
Best Actor/Actress/Director/etc are obviously out for an AI production since those roles simply do not exist.
Similar with Best Visual Effects, I doubt AI generated films qualify.
That leaves us with categories that rate the whole movie (Best Picture, Best International Feature Film etc), sound-related categories (Best Original Score, Original Song, Sound) and maybe Best Cinematography. I doubt the first category is in reach. Video Generation will be good enough in five years. But editing? Screenwriting? Sound Design?
My bet would be on the first AI-related Oscar to be for an AI generated original score or original song, and that no other AI wins Oscars within five years.
Unless we go by a much wider definition of "entirely AI generated" that would allow significant human intervention and supervision. But the more humans are involved the less it has any claim to being "entirely AI". Most AI-generated trailers or the Balenciaga-Potter-style videos still require a lot of human work
I have done quite a bit with AI generated audio/sound/music.
At some point in the process, the end result feels like your own and the models were used to create material for the end work.
At some point, using AI in the creative process will be such a given that it is left unsaid.
I would assume the screen play next year that wins the Oscar will have been helped with the aid of a language model. I can't imagine a writer not using a language model to riff on ideas. The delusional idea here is the prompt "write an Oscar winning screenplay" and that somehow that is all there is going to the creative process.
We are about six years into transformer models. By now we can get transformers to write coherent short stories, and you can get to novel lengths with very careful iterative prompting (e.g. let the AI generate an outline, then chapter summaries, consistency notes, world building, then generate the chapters). But to get anything approaching a good story you still need a lot of manual intervention at all steps of the process. LLMs go off the rail, get pacing completely wrong and demonstrate gaping holes in their understanding of the real world. Progress on new models is mostly focused in other directions, with better storytelling a byproduct. I doubt we get to "best screenplay" level of writing in five years.
Best Actor/Actress/Director/etc are obviously out for an AI production since those roles simply do not exist.
Similar with Best Visual Effects, I doubt AI generated films qualify.
That leaves us with categories that rate the whole movie (Best Picture, Best International Feature Film etc), sound-related categories (Best Original Score, Original Song, Sound) and maybe Best Cinematography. I doubt the first category is in reach. Video Generation will be good enough in five years. But editing? Screenwriting? Sound Design?
My bet would be on the first AI-related Oscar to be for an AI generated original score or original song, and that no other AI wins Oscars within five years.
Unless we go by a much wider definition of "entirely AI generated" that would allow significant human intervention and supervision. But the more humans are involved the less it has any claim to being "entirely AI". Most AI-generated trailers or the Balenciaga-Potter-style videos still require a lot of human work