OpenAI Sora's first short film – "Air Head," created by shy kids [video] | Hacker News Reader

OpenAI Sora's first short film – "Air Head," created by shy kids [video]

youtube.com

22 pointsandersource26 comments

Nice solution to making a film when you can't have a main character who is visibly the same person in multiple scenes.

Even then, the balloon clearly changes between scenes (and it’s still leaving heavily on slo-mo).

If you can’t nail consistency, you’re dead in the water. This js why I’m a bit skeptical that this is going to overhaul the entertainment industry any time soon.

I don't think it will overhaul anything, but I'm excited to work as a hobbyist director at home using similar tools. Animatediff is a really cool start but its still too time consuming for me atm.

This is a known problem in generative workflows for AI vids, but solvable. Midjourney recently introduced a feature that does this for stills, and controlnets available for the comfyui ecosystem also can partially solve this, albeit with some hassle. I'm pretty sure if not OpenAI themselves others will follow with their foundation models.

It was a good idea to put a yellow balloon as the main subject and make it the only colored bit of the video, so that the eye focuses on it and not on the horribly deformed backgrounds that violate Euclidean space in just so many ways.

> and make it the only colored bit of the video

> the horribly deformed backgrounds that violate Euclidean space in just so many ways

You and I watched very different videos, it seems. Neither of those things is true in the video I just watched.

The prospects of the future are so intense that despite me being a lifelong tech junkie, I cannot brush away this dizzying underlying anxiety that arises when I see glimpses of what is to come.

I've been personally feeling like we're opening Pandora's box. Everyone's all starry eyed about LLMs & other AIs, we're shoving it into any product we can find, looking to displace entire industries overnight. The models aren't good enough right now for anyone to be seriously worried, but once genAI is out of the bottle, there's no putting it back. And based on the entire history of humankind, I have 0 faith whatsoever we're going about this responsibly.

Add to that the fact that there's perverse profit incentives around every corner, and I personally think we're about to shoot ourselves in our collective foot, not realizing until it's too late that we probably should have moved more slowly and focused less on capitalist goals & more on collective good.

But AI bros absolutely hate being told to slow down with this tech, so I guess I'll just resign myself to feeling like Cassandra and hoping I'm long dead before we reap what we're currently sowing.

There's a perverse beauty to it - we've collectively beaten the biosphere into submission. Explored the unknowns, commodified everything.

And then, out of our ultimate triumph of order, safety, and control of the natural world comes life! That we can't understand, control, or even necessarily get along with.

Full circle!

Sora is not available, this looks like an attention grab by an app without relation to OpenAI trying to get downloads by false association. Same thing as a year ago with all the ChatGPT iOS apps that came out before OpenAI released theirs.

What makes you think OpenAI hasn't partnered with select studios and creators to use Sora?

It's Sora. https://openai.com/blog/sora-first-impressions

Yes, this video is made with Sora - but this link leads to a youtube video that claims it was created with "KaraVideo.AI's (powered by Sora)" and claims it has "Sora early-access" - which is dishonest and misleading.

Can we replace the link with the actual blog post that doesn't include/promote KaraVideo? https://openai.com/blog/sora-first-impressions

There is a watermark that the video is made by another platform that uses Sora. Are there any descriptions on which materials are by Sora or other software like for the voice?

I like it. However the best bike wheel shot was one where it's going the wrong way?

Wheels appearing to spin the wrong way is actually a very common thing in video because of frame temporal aliasing.

https://en.wikipedia.org/wiki/Wagon-wheel_effect

Hella cuts. More like a montage with VO (for now)

It's a mistake to call this "Sora's" film, any more than a photograph is "Canon's" photo. Sora is just another tool here.

I really dislike how much people have been anthropomorphizing AI tools, like these things are just pumping out content by themselves.

Up next: The Automated YouTube

I think if the Canon camera was able to create all the photos in a black box, regardless of the photographer's skill and surroundings, we would have to give the camera more credit than the so-called photographer.

You think this short film was created "regardless of skill", i.e. by someone clicking a "generate" button?

No, I don't. But pretty close. They did very little work that could be called film making.

They have prompt engineering skill. If you can't do your creative job without AI, you can't do the job. The job you are doing is prompt engineering.

Also writing, editing, sound engineering, voiceover work. I’m not saying it’s great art but there are a lot of filmmaking disciplines on display here.

> They have prompt engineering skill. If you can't do your creative job without AI, you can't do the job. The job you are doing is prompt engineering.

In contrast to what you said above, below is how the team describes themselves:

> shy kids, a team of multi-faceted artists, began as three friends from toronto, canada and has now evolved into a small but versatile production company. they create most of their projects from scratch, proficient in various aspects such as writing, directing, producing, shooting, editing, animating, VFX, and composing music. they are known for their vibrant animation, dubbed “punk-rock pixar”. their work has garnered emmy nominations and been long-listed for the oscars. currently, shy kids is involved in producing series for major entertainment platforms like Disney, HBO, AMC, and Netflix.

Maybe you can submit a pull request to their website to change that text to "we are prompt engineers who click buttons on websites".

Lol. Perhaps I will...

Reply on news.ycombinator.com