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

311–320 of 916 posts

Re: Sora 2

#311

Post by cubefox »

viccis wrote: post #85
minimaxir wrote: post #79

Earlier quoted context omitted.

The same could be said about software, and it's safe to say that open-source software making complex workflows easier and more efficient is a net good. Making better tools is better for everyone: the median usage of those tools downstream is a separate issue.

If you're comparing how art is evaluated to how software is evaluated then it sounds like you only understand one or the other.

Indeed. Art is partially evaluated by how impressive it is. That's why posting AI images on social media won't yield a lot of likes anymore. People have gotten used to images being easy to create, so they aren't seen as valuable anymore. The same will be true for videos.

AI pictures today are much less impressive than Dall-E 2 pictures were a few years ago, despite the fact that the models are much better nowadays. Currently AI videos can still be impressive, but this will quickly become a thing of the past.

Then people will move from trying to create art to creating "content". That is, non-artistic slop. Advertisements. Porn. Meme jokes. Click bait. Rage bait. Propaganda. Etc.

Re: Sora 2

#312

Post by fluoridation »

marcofloriano wrote: post #217

Every AI video demonstration is always about funny stuff and fancy situations. We never see videos on art, history, literature, poetry, religion (imagine building a video about the moment Jesus was born) ... ducks in a race !? Come on ... So much visual power, yet so little soul power. We are dying.

What do you imagine a generated video about poetry would be?

>Every AI video demonstration is always about funny stuff and fancy situations.

The thing about AI slop is that by its very nature, unless it's heavily reined in by a human, it's invariably lowest common denominator garbage. It very likely will generate something you yourself could think of within the first five seconds of hearing the prompt, not some very clever take on it, so it can only work as a placeholder (AI as a replacement of stock images is great, for example) or to add background detail where it won't call attention to itself and its genericity.

>imagine building a video about the moment Jesus was born

Given there are multiple paintings on the subject, I very much doubt no one has generated something like that already.

Re: Sora 2

#313

Post by bob1029 »

polishdude20 wrote: post #246

There's something about the faces that looks completely off to me. I think it's the way the mouth and whole face moves when they talk.

It's the inaccuracy of things like shadows, sub-surface scattering and specular highlights. I think the shadow inaccuracy is what the human visual system is most sensitive to.

These LLMs might make content that looks initially impressive but they are absolutely not performing physically based rendering or have any awareness of the lighting arrangement in these scenes. There are a lot of things they get right, but you only have to screw up one small element to throw the whole thing off.

I am willing to bet that Unreal Engine 5 will continue to produce more realistic human faces than OAI ever can with these types of models. You cannot beat the effects of actually running raytracing in a PBR pipeline.

Re: Sora 2

#314

Post by ronsor »

haolez wrote: post #256
ronsor wrote: post #184

Earlier quoted context omitted.

> (I'm ignoring copyright aspects, of course, because those are too boring :D) You must understand that infinite copyright is the author's right, and AI companies must be sued for 50 trillion dollars.

Come on. This is just a fun thought exercise. I'm not suggesting creating a startup around this.

I was trying my hand at satire; but I understand that many people now genuinely hold such extreme views.

Re: Sora 2

#317

Post by cubefox »

FergusArgyll wrote: post #142
rhetocj23 wrote: post #83

Earlier quoted context omitted.

This is absolutely horrible. People need to be exposed to what is real. Not more artificial stuff. I think this is the point at which humanity will finally puke and reject this crap. Just because a small segment of people like it doesnt mean the mass majority will.

I personally love Monet, he's not for everyone, I know, but I'm sure you can find some art you appreciate

You probably don't personally love AI generated impressionist content.

Re: Sora 2

#318

Post by Computer0 »

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…

Are users of the $20 tier really going to have to deal with that obnoxious bouncing watermark I wonder? The previous watermark could be cropped, but I often didn't feel the need to as I use it for fun, but that would make me not want to show anyone.

Re: Sora 2

#319

Post by cubefox »

thorum wrote: post #92
busymom0 wrote: post #74

Earlier quoted context omitted.

> People are doing cool things with it Things are cool because they are unique, very hard to create, and require creativity. When those things become cheap commodities, they are no longer cool.

I would argue that we just get pickier and more sensitive to slop. When everyone can make a movie, the standard for a good movie will be higher. Many current Hollywood films wouldn’t make the cut. But maybe some kid in Nigeria makes the greatest film of all time.

By that logic, some kid in Nigeria could have written the greatest book of all time. At least by commonly accepted measures, that didn't happen.

Re: Sora 2

#320

Post by SeanAnderson »

iLoveOncall wrote: post #299
SeanAnderson wrote: post #250

Sheeeeeeeeeeesh. That was so impressive. I had to go back to the start and confirm it said "Everything you're about to see is Sora 2" when I saw Sam do that intro. I thought there was a prologue that was native film before getting to the generated content.

I'm sorry but that's a gross exageration. If any of this was real film then I'd start a gofundme page for OpenAI to get better video production equipment and team because that would be laughably bad. If anything, it looks a lot worse than a lot of AI-generated videos I've seen in the past, despite being a tech demo with carefully curated shots. Veo 3 just blows this out of the water for example.

It's not an exaggeration to me? I literally stopped the video and went back to the start and re-read. You're more than welcome to speak about your opinions and experiences, but I'm speaking about mine.

I'm over here thinking, "It felt like just yesterday I was laughing at trippy, incoherent videos of Will Smith eating spaghetti."

I love the progress we're making. I love the competition between big companies trying to make the most appealing product demos. I love not knowing what the tech world is going to look like in six months. I love not thinking, "Man. The Internet was a cool invention to have grown up in, but now all tech is mundane and extractive." Every time I see AI progress I'm filled with childlike wonder that I thought was gone for good.

I don't know if this represent SOTA for video generation. I don't care. In that moment I found it impressive and was commenting specifically on the joy I experienced watching the video. I find it frustrating to have that joy met with such negativity.

Post reply on HN