I love the statement in the conclusion.
Curation is something we intrinsically favor over engagement algorithms. Noisy is easy to quantify, but greatness is not. Greatness might have a lag in engagement metrics while folks read or watch the material. It might provoke consideration, instead of reaction.
Often we need seasons of production in order to calibrate our selection criteria, and hopefully this season of booming generation leads to a very rich new opportunity to curate great things to elevate from the noise.
I spend a lot of my time researching live-action anime[0][1], and there's an important thing to learn from Japanese animators: sometimes an animation style may seem technically lacking, but visually stunning.
When animator Ken Arto was on the Trash Taste podcast he mentioned how Disney had the resources to perfect the animation, while in Japan they had to achieve more with less.
This basically shifts the "what is good animation" discussion in ways that are not as clear from looking at the stats.
[0] https://blog.nestful.app/p/ways-to-use-nestful-outlining-ani...
These kinds of perspectives are often found and parroted in perceived 'elite' circles. It's no wonder the author works in Epic Games, a place in which one would need high technical chops to work there.
It's also no wonder why such people get disconnected from some realities on the ground. Sure on paper people do want higher quality things but they don't even know what those are. Most people have low-brow tastes; they'd take a cheaper and well-marketed thing over a 1% improvement.
Japan didn't need to compete on the same ladder for success, it needed to mix various elements of what they're good at to achieve it's own success.
The author did some very cool work with Raylib interpolating between animations to make transitions more natural. I remember being blown away at how realistic it looked from the videos he posted in the Discord. Glad to see he's still pushing the boundaries on what's possible with quality animation. And of course Cello rocks!
Something I keep seeing is that modern ML makes for some really cool and impressive tech demos in the creative field, but is not productionizable due to a lack of creative control.
Namely, anything generating music / video / images - tweaking the output is not workable.
Some notable exceptions are when you need stock art for a blog post (no need for creative control), Adobe's recolorization tool (lots of control built in), and a couple more things here and there.
I don't know how it is for 3D assets or rigged model animation (as per the article), never worked with them. I'd be curious to hear about successful applications, maybe there's a pattern.
Probably accurate for videos and music. Videos because there’s going to be just too many things to correct to make it time efficient. Music because music just needs to be excellent or it’s trash. That is for high quality art of course. You can ship filler garbage for lots of things.
2D art has a lot of strong tooling though. If you’re actually trying to use AI art tooling, you won’t be just dropping a prompt and hoping for the best. You will be using a workflow graph and carefully iterating on the same image with controlled seeds and then specific areas for inpainting.
We are at an awkward inflection point where we have great tooling for the last generation of models like SDXL, but haven’t really made them ready for the current gen of models (Flux) which are substantially better. But it’s basically an inevitability on the order of months.
Even with the relatively strong tooling for 2D art it's still very difficult to push the generated image in novel directions though, hence the heavy reliance on LoRAs trained on prior examples. There doesn't seem to be an answer to "how would you create [artists] style with AI" that doesn't require [artist] to already exist so you can throw their life's work into a blender and make a model that copies it.
I've found this to be observable in practice - I follow hundreds of artists who I could reliably name by seeing an example of their work, even if they're only amateurs, but I find that AI art just blurs together into a samey mush with nothing to distinguish the person at the wheel from anyone else using the same models. The tool speaks much louder than the person supposedly directing it, which isn't the case with say Photoshop, Clip Studio or Blender.
Shrug. That’s a very different goal. Yes, if you want to leverage a different style your best bet is to train a Lora off a dozen images in that style.
Art made by unskilled randos is always going to blur together. But the question I feel we’re discussing here is whether a dedicated artist can use them for production grade content. And the answer is yes.
The generated artwork will initially displace clipart/stock footage and then illustrators and graphic designers.
The last 2 can have tremendous talent but the society at large isn’t that sensitive to the higher quality output.
Something I realized about AI is that an AI that generates "art" be it text, image, animation, video, photography, etc., is cool. The product it generates, however, is not.
It's very cool that we have a technology that can generate video, but what's cool is the tech, not the video. It doesn't matter if it's a man eating spaghetti or a woman walking in front of dozens of reflections. The tech is cool, the video is not. It could be ANY video and just the fact AI can generate is cool. But nobody likes a video that is generated by AI.
A very cool technology to produce products that nobody wants.
That's an over simplification I think. If you're only generating a video because 'I can oooh AI' - then of course no one wants it. If you treat the tools as what they are, Tools - then people may want it.
No one really cares about a tech demo, but if generative tools help you make a cool music video to an awesome song? People will want it.
Well, as long as they aren't put off by a regressive stigma against new tool at least.
Are there any valid reasons people might not like this or is it only "regressive stigma?"
If you used AI to make something awesome, even if I liked it, I'd feel scammed if it wasn't clearly labelled as AI, and if it was clearly labelled as AI I wouldn't even look at it.
The problem in your example is that you wouldn’t think a picture of a man eating spaghetti taken by a real person would be cool.
You may feel different if it’s, say, art assets in your new favorite video game, frames of a show, or supplementary art assets in some sort of media.
> A very cool technology to produce products that nobody wants.
creative power without control is like a rocket with no navigation—sure, you'll launch, but who knows where you'll crash!
> or a woman walking in front of dozens of reflections
A lot of people will not notice the missing reflections and because of this our gatekeepers to quality will disappear.
While I am in the same camp as you, there is one exception: Music. Especially music with lyrics (like suno.com) - Although I know that it's not created by humans, the music created by Suno is still very listenable and it evokes feelings just like any other piece of music does. Especially if I am on a playlist and doing something else and the songs just progress into the unknown. Even when I am in a more conscious state - i.e. creating my own songs in Suno, the end result is so good that I can listen to it over and over again. Especially those ones that I create for special events (like mocking a friend's passing phase of communism and reverting back to capitalism).
In my opinion, Suno is good for making really funny songs, but not for making really moving songs. Examples of songs that make me chuckle that I've had it do:
A Bluegrass song about how much fun it is to punch holes in drywall like a karate master.
A post-punk/hardcore song about the taste of the mud and rocks at the bottom of a mountain stream in the newly formed mountains of Oklahoma.
A hair band power ballad about white dad sneakers.
But for "serious" songs, the end result sounds like generic muzak you might hear in the background at Wal-Mart.
appreciate your position but mine is that everything out of suno sounds like copycat dog water.
Makes sense that GP appreciates the taste of dog water when they’re mocking their friends for having had values (friends whom likely gave up their values to stop being mocked)
Yes, it turns out there's more to creating good art than simulating the mechanics and technique of good artists. The human factor actually matters, and that factor can't be extrapolated from the data in the model itself. In essence it's a lossy compression problem.
It is technically interesting, and a lot of what it creates does have its own aesthetic appeal just because of how uncanny it can get, particularly in a photorealistic format. It's like looking at the product of an alien mind, or an alternate reality. But as an expression of actual human creative potential and directed intent I think it will always fall short of the tools we already have. They require skilled human beings who require paychecks and sustenance and sleep and toilets, and sometimes form unions, and unfortunately that's the problem AI is being deployed to solve in the hope that "extruded AI art product" is good enough to make a profit from.
Seems like this site is getting hugged to death right now
I haven't checked, but I think some of the videos on the page might be served directly from the server.
Edit: Wow! they are loaded directly from the server where I assume no cdn is involved. And what's even worse they're not lazy loaded. No wonder why it cannot handle a little bit of traffic.