Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Sora 2

openai.com

Post reply on HN
Print view View original

401–410 of 916 posts

Re: Sora 2

#401

Post by the_duke »

password54321 wrote: post #400
echelon wrote: post #339

Earlier quoted context omitted.

I posit this is the real story. OpenAI did not stealthily release Sora 2 to the image and video ELO ranking leaderboards ahead of time as is now somewhat tradition. This model is probably designed to run fast and cheap as a social play. Emphasis on putting you and your friends into popular franchises and IPs. OpenAI probably has a totally different model for their Hollywood-grade VFX. One that's too expensive to offe…

Touch grass. This is nothing but cringe that I wouldn't wish upon children.

Kids are going to absolutely love this.

Re: Sora 2

#402

Post by Tiktaalik »

stan_kirdey wrote: post #129

That could totally power next generation of green-screen techs. Generative actors may not find favorable response in the audiences; but SFX, decor, extras, environments that react to actors' actions - amazing potential.

There's less here than you think. Video games have already been procedurally generating environment art for quite some time, and film/tv are already leveraging that with giant screens that use Unreal Engine to create the backgrounds.

AI could be helpful here, but it's not clear that it is required or an improvement.

Re: Sora 2

#403

Post by shubb »

larodi wrote: post #262
saguntum wrote: post #205

I wonder if they're going to license this to brands for heavily personalized advertisement. Imagine being able to see videos of yourself wearing clothes you're buying online before you actually place the order, instead of viewing them on a model. If they got the generation "live" enough, imagine walking past a mirror in a department store and seeing yourself in different clothes. Wild times.

its called Virtual Try On (VTO) and there are plenty of models going there for static gfx, it is very reasonable to expect soon emerge those for video VTO.

Accurate virtual try on however is quite difficult, and users will quickly learn to distrust platforms that just generate something that"looks right".

You can prompt with a normal size 8 dress and "kim jungle un wearing a dress" and it will show you something that doesn't help you understand whether that dress would fit or not. You can ask for a tube dress and it will usually give him a big bust to hold it up. It's not useful for the purpose of visualing fit.

It will definitely be used for such just like image models already are for cheap tenu clothes, and our onions shopping experience will get worse.

Maybe this needs purpose built models like vibe-net or maybe you cab train a general purpose model to do it, but if they were spending the effort necessary to do so they'd be calling it out.

Re: Sora 2

#404

Post by rvz »

the_duke wrote: post #273

I haven't seen comments regarding a big factor here: It seems like OpenAI is trying to turn Sora into a social network - TikTok but AI. The webapp is heavily geared towards consumption, with a feed as the entry point, liking and commenting for posts, and user profiles having a prominent role. The creation aspect seems about as important as on Instagram, TikTok etc - easily available, but not the primary focus. Genera…

I bet xAI and X will likely relaunch Vine with AI videos as a competitor to Sora 2.

Re: Sora 2

#406

Post by squidsoup »

echelon wrote: post #229

I'm a software engineer and hobbyist actor/director. My friends are in the film industry and are in IATSE and SAG-AFTRA. I've made photons-on-glass films for decades, and I frequently film stuff with my friends for festivals. I love this AI video technology. Here are some of the films my friends and I have been making with AI. These are not "prompted", but instead use a lot of hand animation, rotoscoping, and human v…

Creative people with ambition and limited resources make good things today without this technology. All this does is accelerate the rate at which low quality "content" is produced by people that have no interest in learning a craft, without attribution and without compensation for the people that have made the effort and whose works train these models.

Re: Sora 2

#407

Post by nycdatasci »

What makes TikTok fun is seeing actual people do crazy stuff. Sora 2 could synthesize someone hitting five full-court shots in a row, but it wouldn’t be inspiring or engaging. How will this be different than music-generating AI like Suno, which doesn't have widespread adoption despite incredible capabilities?

Re: Sora 2

#408

Post by rhetocj23 »

bflesch wrote: post #357
minimaxir wrote: post #7

OpenAI apparently assumes that the primary users of Sora 2/the Sora app will be Gen Z, especially with the demo examples shown in the livestream. If they are trying to pull users from TikTok with this, it won't work: there's some nuance to Gen Z interests than being quirky and random, and if they did indeed pull users from TikTok then ByteDance could easily include their own image/video generators. Sora 2 itself as a…

Good point. I think OpenAI lacks the cultural understanding that tiktok is providing their users not only with entertainment but also social things like trends, reviews, gossip, self-expression. These aspects are not included in the sora experience.

This is going to sound crass but idc - OAI is just full of geeks, when what is needed is people who are more akin to hippies - thats pretty much what Apple was in the early days.

Its no use building technology when its not married with the humanities and liberal arts.

Re: Sora 2

#409

Post by dolebirchwood »

This makes me less excited about the future of video, not more.

It's technically impressive, but all so very soulless.

When everything fake feels real, will everything real feel fake?

Re: Sora 2

#410

Post by rhetocj23 »

squidsoup wrote: post #406
echelon wrote: post #229

I'm a software engineer and hobbyist actor/director. My friends are in the film industry and are in IATSE and SAG-AFTRA. I've made photons-on-glass films for decades, and I frequently film stuff with my friends for festivals. I love this AI video technology. Here are some of the films my friends and I have been making with AI. These are not "prompted", but instead use a lot of hand animation, rotoscoping, and human v…

Creative people with ambition and limited resources make good things today without this technology. All this does is accelerate the rate at which low quality "content" is produced by people that have no interest in learning a craft, without attribution and without compensation for the people that have made the effort and whose works train these models.

Precisely.

I have a really big problem with letting low quality stuff infest into the species.

Post reply on HN