aenvoker

I find the test of “show me AI art that isn’t a rehash” to be a bit pernicious because I have a hard time coming up with any recent examples of non-AI art that I could lie about and claim was AI generated and you’d respond “OK, that AI art is not a rehash.” If everything is a rehash, what are we arguing about?

Instead, my goto examples of high quality creative works that are clearly grounded on AI tools is https://andreaciulu.com/projects/ and https://youtu.be/fyZhC2TXgcs

“model that doesn't possess any agency” implies a model acting on its own. Like a paint brush flying around the room painting the walls with no people around. Or a computer spontaneously turning on and publishing images to an online gallery with no on operating it. The creative process of using AI tools is one of exploration, pipelining and editing.

Commentators strawman image generators as magic paintbrush machine guns. But, the process of using them intentionally is more like nature/street photography. You can’t walk into the forest and declare “I will photograph an owl fighting a fox over a mouse today.” But, you can develop a sense for where to go to find good shots. Now imagine instead of walking in a forest, you can pop anywhere in the multiverse. You don’t know exactly what to expect each time. And, the space is infinite in many dimensions. But, you can still learn to navigate.

And, that’s just the process of direct prompting. One step past that gets you configuring pipelines of highly varied generative editing tools like https://www.runcomfy.com/comfyui-workflows/comfyui-product-r...

From there you get into integrations with 2D painting/3D modeling tools and video editors. Generating references which are used to generate style sheets, then characters, sets, blockings, dialog, motion, base renders, upscales, edits.

“Make me a pretty picture. Masterpiece, best quality” is easy to make fun of. And, it’s common because so easy many people are playing around with it. But, it’s tiresome to find all of the discussion assumes that’s all there can ever be.

I was a content moderator for Midjourney for many years. I’ve seen more AI-generated debauchery than I could ever hope to convey. I also saw 100,000x more joy, creativity, beauty and expression in that same community.

Unfortunately, millions of people playing around having low-key fun making low-effort art is a boring headline. One person getting caught trying to be gross with AI drives a million times more media engagement than a million people trying to be creative for once. So, guess which one the media focuses on.

Web devs gonna focus on web dev things. But, I’m more interested in the works of artists like https://www.rooperainisto.com/ and https://andreaciulu.com/ who embraced the glitchy aesthetic of early AI image generation and made it integral to their art instead of fighting it.

As a clanker-apologist I gotta say this is the strawman of AI slop brought to life. Letting the machine do literally everything without even supervision and see how it turns out? No surprise at all the results are so bad.

My fav AI video is still Post-Scarcity Blues from a year ago https://www.youtube.com/watch?v=q_t3h2AZ0KY There have been others I've enjoyed since then. But, that one stands out in memory. Work warning: it is occasionally just a bit spicy.

Midjourney is not actually an AI imagery company. It's a research lab that happened to do AI images first.

I talked more about it here: https://news.ycombinator.com/item?id=48588293

I know the founder through a lot of light discussion. If lots of for-profit alternatives become serious competition, he'd rather move on to something else than take on VC money. His POV is that he already has more money than he ever wanted. All he wants now is to do fun research.

Midjourney is not actually an AI imagery company. It's a research lab that happened to do AI images first.

I talked more about it here: https://news.ycombinator.com/item?id=48588293

Midjourney is not actually an AI imagery company. It's a research lab that happened to do AI images first.

The founder is a hardware guy who made enough money to retire young off of the sale of his company, Leap Motion. But, he decided what he really wanted to do was cool research with cool people. So, he started Midjourney. The goal for the AI image generator was to be cool research, pay for itself, and grow the lab. It ended up making far more money than ever expected.

I was a Discord mod for Midjourney when it was still in private beta. I got to participate in some of the discussions of "WTH are we doing and how should we do it?" DavidH is very much a smart hippie idealist. He isn't really motivated by even more money beyond how it enables more fun research. MJ actively refused investment. And, actively refused partnerships that would make them money but wouldn't help build the community or the lab.

So, put together: I can totally see how this looks weird from the outside. But, having spent a few years peeking inside, I'm only surprised it took so long for them to branch out like this.

Ecma Technical Committee 57: Standardization of the High-Level Shading Language devblogs.microsoft.com

7 aenvoker 9 months ago 1 comments

Comment · 9 months ago

Not on the VC path. Not even on the max-profit path. Just on the "Have fun doing cool research" path.

I was a mod on MJ for its first few years and got to know MJ's founder through discussions there. He already had "enough" money for himself from his prior sale of Leap Motion to do whatever he wanted. And, he decided what he wanted was to do cool research with fun people. So, he started MJ. Now he has far more money than before and what he wants to do with it is to have more fun doing more cool research.

I rather enjoyed some of these

https://reddit.com/r/singularity/comments/1lq299r/postscarci...

https://reddit.com/r/midjourney/comments/1o6ickx/dreaming_on...

https://reddit.com/r/midjourney/comments/1n6mzig/how_to_buil...

https://reddit.com/r/aivideo/comments/1nwdjdn/the_perfect_bo...

https://reddit.com/r/aivideo/comments/1m8a9wz/pinkington_rop...

https://reddit.com/r/aivideo/comments/1n52kut/derek_the_agin...

https://reddit.com/r/midjourney/comments/1muwyah/still_here_...

https://reddit.com/r/DefendingAIArt/comments/1mttoi4/my_not_...

Having spent a lot of time talking to engineers working on AI tools, I find the idea that AI is intended to replace creativity comes entirely from internet comment sections and not from the engineers working on AI.

Having used lots of different image generators, so far the only one with a sense of humor has been Dall-e3.

https://www.reddit.com/r/dalle2/s/khb5XuNFdl

There’s probably some sort of connection to ChatGPT in there. But, I don’t know enough about how it works.

SD has a very primitive conceptual model. Basically “bag of words nudging pixels around for a while”. Words near each other influence each other. But, there’s nearly no understanding of grammar.

Midjourney is similar with text prompts. But, with image prompts it is able to understand content separately from style. You can give it a photo of two people and it can return many images of recognizable approximations of those people in different poses.

SD can only start from pixels, blur and deblur those pixels in place.

MJ image prompts probably works via image-to-tokens added on to your text-to-tokens-to-image.