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

Sora: Creating video from text

openai.com

Post reply on HN
Print view View original

321–330 of 1001 posts

Re: Sora: Creating video from text

#321

Post by CSMastermind »

mbil wrote: post #41
hansonkd wrote: post #26

Countdown to when studios licensing this for "unlimited" episodes of your favorite series. There was Seinfeld "Nothing, Forever" AI parody, but once the models improve enough and are cheap enough to deploy, studios will license their content for real and just have endless seasons. Or even custom episodes. Imagine if every episode of a TV show was unique to the viewer.

I imagine it's not long before we see hyper-targeted commercials where the actors look like us, live in our city, etc.

There was some monitoring company that used to have creepy web ads that would show the actual company you worked at in the ads.

If anything it was a turn off and I was confused how they knew where I worked.

Re: Sora: Creating video from text

#322

Post by Workaccount2 »

cboswel1 wrote: post #262

Who owns a person’s likeness? Now that we’re approaching text to video of a quality that could fool an average person, won’t this just open a whole new can of worms if the training models are replicating celebrities? The ambiguity around copyright when something on paper is in the style of seems to fall into an entirely separate category than making AI generated videos of actual people without their consent. Will peo…

$100 on the table that studios create new celebrities that they own the rights too.

Re: Sora: Creating video from text

#323

Post by qiller »

jenny91 wrote: post #253

Absolutely insane. It's very odd where the glitches happen. Did anyone else notice in the "stylish woman ... Tokyo" clip how her legs skip-hop and then cross at 0:30 in a physically impossible way. Everything else about the clip seems so realistic, yet this is where it trips up ?

The construction scene has people appearing out of thin air, changing jacket colors, and in general weird things happening

Re: Sora: Creating video from text

#325

Post by rareitem »

I used to think a few years ago that virtual reality/ai projects such as the mataverse wouldn't amount to anything big. I even thought of them ridiculous. Even recently, I thought that GPT's and ai generated images would be the pinnacle of what this new ai wave would amount to. I just keep getting baffled.

Re: Sora: Creating video from text

#327

Post by gwern »

throwitaway222 wrote: post #287

https://openai.com/sora?video=cat-on-bed Even though many things are super impressive, there is a lot of uncanny valley happening here.

Cats are like hands: they are hilariously hard for generative models and then after thinking about it, you realize that cats/hands really are hard. I mean, look at photos of a black cat curled up where it might have its paws sticking out at any angle from anywhere from a solid black void. How the heck do you learn that?

Re: Sora: Creating video from text

#329

Post by mbm »

hooande wrote: post #303

This really seems like "DALL-E", but for videos. I can make cool/funny videos for my friends, but after a while the novelty wears off. All of the AI generated media has this quality where I can immediately tell that it's ai, and that becomes my dominant thought. I see these things on social media and think "oh, another ai pic" and keep scrolling. I've yet to be confused about whether something is ai generated or real…

Yeah, you really have to fast-forward 5 to 10 years. The first cars or airplanes didn't run particularly well either. Soon enough, we won't be able to tell.

Re: Sora: Creating video from text

#330

Post by neilk »

jijijijij wrote: post #237
thepasswordis wrote: post #100

OpenAI demonstrating the size of their moat. How many multi-million-dollar funded startups did this just absolutely obsolete? This is so, so, so much better than every other generative video AI we've seen. Most of those were basically a still image with a very slowly moving background. This is not that. Sam is probably going to get his $7T if he keeps this up, and when he does everybody else will be locked out foreve…

I predict, this "AI" content generation will eat itself at last. It will outcompete the low-effort "content" industry as is. Then inevitably completely devalue this sort of "product". Because it will never get to 100% of the real thing, the "AI" content craze will ultimately implode. I bet we won't get AGI as a progression of this very technology. The impression of "usefulness" will end when "AI" is starting to drink…

I think this analysis is flawed. New technologies are usually bad at substituting for things that already exist. It's 100% true this will not substitute for the existing genre of film and video.

New technologies change the economics of how we satisfy our needs.

When search engines became good, many pundits confidently predicted Google would never replace librarians or libraries. It didn't. It shifted our relationship to knowledge; instead of having to employ an expert in looking things up, we all had to become experts at sifting through a flood of info.

When the cost of producing art-directed and realistic video goes to zero it's hard to predict what's going to happen. Obviously the era of video = veracity is now over. And you can get the equivalent of Martin Scorsese and a million dollar budget to do the video instructions for a hair dryer. Instead of hunting for a gif to express how you feel, captured from an existing TV show or something, you could create a scene on the fly and attach it to a text message. Or maybe you dispense with text messages altogether. Maybe text is only for talking to computers now.

My personal prediction is that the value of a degree in art history is going to go way, way up, because they'll be the best prompt engineers. And just like desktop publishing spawned legions of amateur typesetters, it will create lots of lore among amateur video creators.

Post reply on HN