The limits of CGI have gotten so good that if you are noticing the CGI, it's b/c someone skimped on the budget for the particular movie where you saw the uncanny valley.

(Of course, excluding the obvious "that guy just knocked down a building!" CGI)

If they're actually hosting it and providing a service, that wouldn't be a bad thing. Ovi's Apache license allows commercial use. They would only be infringing on the license by not publishing it and the original copyright, and would be morally bankrupt by not disclosing the original project and passing it off as their own.

But I also suspect that most of these are indeed SEO scammers, that there's no actual service, and that all payments are pocketed. It might take a few days for the scam to be reported and the site taken down, but it's likely enough to get a few hundred bucks out of it. They'll never be pursued because of where they live, and they can have many of these up in no time, thanks to AI, as you say.

What a sad state of affairs that no "AI" company or government is taking seriously.

OpenAI wowed the world with a video model that was also a script writing, editing, dubbing, foley, and music model.

Kling still has the best proprietary video model, but Sora 2 is so smart that you don't need to edit anything if your target is social.

I don't see how Runway, Pika, or the rest of the purely foundation video model startups survive against the giants and the incredible open source Chinese models. They've got to be sweating bullets right now.

Everyone's also sleeping on xAI's high quality and insanely fast video model (10 second generations) that they're giving away completely for free without watermarks.

It's a relatively well defined measure of success though: a movie which is popular and high-grossing.

I can see soap-opera-style, video-manga becoming a thing, where you get 5x20 minute episodes a week, and an ai-generated 30 min super cut of the week's "events" every saturday. What if DragonBallZ, but new episodes drop every morning before your morning commute? And for $30 a month you can do a choose your own adventure where at the end of each episode, the story splits off and you get an alternate history version, but conveniently rejoins the main story by the end of the week.

Three years ago we had a live streaming autogen-seinfeld twitch stream; some kind of coherent story telling via AI doesn't seem beyond reach today, the tools just haven't fully matured yet.

People can daydream for free now.

Never is a long time. The resistance will erode.

This is defo true until the moment it gets so good people can't tell.

I read a while ago that big scientific ideas take about 50 years to be accepted. Which basically means they are never accepted. The people who disagree just get old and die.

Younger generation who grow up with AI will just think it’s normal, like we think being connected to the internet via a rectangle you keep in your pocket is normal.

New things will be possible that aren’t today. You’ll be able to pick the stars in your movie. The home base can be your childhood home. Your unrequited love can be virtually fulfilled. Etc etc.

honestly i think there is a future in this.. "hey AI model, generate a movie from this book" may indeed be better than letting a director mangle it into something meant to sell tickets. most book -> movie adaptations are not setting a high bar

Photos, movies, animation, recorded music, computer games, CGI props in movies, CGI characters in movies were all denounced as "not real art", until good counter-examples appeared.

AI is but a tool; if there is an artist using them, real art can be created, as with any other tool.

I enjoyed just about everything in here: https://andymasley.substack.com/p/a-ton-of-ai-images-ive-mad...

That and control, there's an enormous gap between letting a model just YOLO a prompt and fill in 95% of the details with vaguely plausible whatever, compared to getting a model to execute on a meticulously planned out shot with every last detail in order.