Google Veo 3 fails, week 2: fail harder — Shokunin Studio spins the gacha
Google Veo 3 fails, week 2: fail harder — Shokunin Studio spins the gacha
This is our second week of running Google’s Veo 3 video generator through its paces. So when you’ve read this, go and watch the worked examples. And check out last week’s. [YouTube] Our good friend…
don't just read the text - watch the video, it's the point of the exercise. And I worked all weekend on it. Thank you.
You're viewing a single thread.
One of the YouTube comments was actually kind of interesting in trying to think through just how wildly you would need to change the creative process in order to allow for the quirks and inadequacies of this "tool". It really does seem like GenAI is worse than useless for any kind of artistic or communicative project. If you have something specific you want to say or you have something specific you want to create the outputs of these tools are not going to be that, no matter how carefully you describe it in the prompt. Not only that, but the underlying process of working in pixels, frames, or tokens natively, rather than as a consequence of trying to create objects, motions, or ideas, means that those outputs are often not even a very useful starting point.
This basically leaves software development and spam as the only two areas I can think of where GenAI has a potential future, because they're the only fields where the output being interpretable by a computer is just as if not more important than whatever its actual contents are.
If computers become capable of mass-producing stuff other computers will like, but many humans won't, this might also lead to a quick decline of algorithm-based search engines, social media feeds etc. (as has been discussed here before, of course).
I started thinking about what kind of story you could tell with these impressive but incoherent bits. It wouldn't be a typical movie, but there's got to be a ton of money willing to back any movie that can claim to be "made with AI".
One would have to start from the technical limitations. The characters are inconsistent, so in order to tell any story one would need something that the technology can deliver at least a high percentage of the time to identify protagonist/antagonist. Perhaps hats in different colours? Or film protagonist and antagonists with green screen and put them in the clips? (That is cheating, but of course they would cheat.)
So what kind of story can you tell? A movie that perhaps has a lot of dream sequences? Or a drug trip? It would be very niche, but again the point would just be to be able to claim "made with AI".
You can film with an actual camera then use video to video to make it look very AI. If you're just grifting, that would be the way to go I think.
Clever. Writing up my pitch to Open ai...
It seems pretty perfect for surrealism where everything is supposed to be illogical and dreamlike.
So what kind of story can you tell? A movie that perhaps has a lot of dream sequences? Or a drug trip?
Maybe something like time travel, because then it might be okay if the protagonists kept changing their appearance to some degree. But even then, there wouldn't be enough consistency, I guess.
I feel like there's got to be a surreal horror movie in there somewhere. Like an AI-assisted Videodrome or something.