AI is going to be our biggest game changer
This discussion has been closed.
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2025 Daz Productions Inc. All Rights Reserved.
Comments
I guess some people just want nice images, no matter what. But most people want to control how their visions look and that is not possible with AI yet, you cannot control the camera angle, clothes etc. what if you wanna create the same character in different environments? not possible, zero control.
its great to get some ideas, but thats it.
The argument they will make is that a person had to type very specific words to get the AI to generate that picture. This was basically the argument that the guy who won the state fair made for why his picture should be called art.
That is my new skill, BTW. I can smooth talk a computer, baby yeah!
Anyway, I also wanted to post a video with some real freakin music in it. The person behind this used lyrics in the song for prompts. He must have added some words for the various intrumental parts of the song, and the images for cowbell are kind of amazing. It turned out alright. It is fitting to the tune and band. However, the one thing I notice in nearly all of these paintings is they have a similar style going on. These do not look out of place from that state fair piece mentioned earlier. So the AI is clearly drawing upon a similar source material.
Almost forgot, the same youtuber has posted several other classic rock/pop songs with videos made this way. Stairway to Heaven...whoa.
OK, that was amazing.
after spending 6 hours rendering this video I Zremeshed the 4Mil+ poly mesh
a planar UV the original texture still fits so future videos will be faster
The problem of reproducable characters is being worked on. One approach this person took is using Textual Inversion on a 3D rendered character that they rendered from 7 different angles/lighting and trained the AI on those images. They could then style transfer and edit this newly learned character:
https://www.reddit.com/r/StableDiffusion/comments/xia53p/textual_inversion_results_trained_on_my_3d/
I almost snarfed at this statement when instantly an image of hedonism bot came to mind -
And yes, I totally agree. I've been keeping eyes on this thread and it honestly just gets more and more disheartening for me. I haven't even opened Daz in weeks and it's pretty much all stemming from the "why bother" feeling I'm getting from all this. Images I see spitting out on AI forums are better (I feel) than anything I can create in Daz/Photoshop...stuff I've spent 40...50...60+ hours on. Yes, I can control the exact pose and exact facial expression and exact camera angle and every little thing about the piece to tell exactly the story I want to tell, but as others have already said, it seems like more and more folks just don't really care about all that when there's a "Robot, make me a picture..." button.
I've also been following the discussions around AI art and I'm not deaf to the mixed feelings about it. On one hand there's angst and dismay; on the other, jubilation over the discovery of a new enabling tool. I can't help but be philosophical about this, that somehow "All this has happened before, and it will happen again" (quote Peter Pan, or Battlestar Galactica).
If we look at history we see that art movements swings to and fro, often in reaction to a novel technique that some new technology brought forth. There are clashes between swings and counter-swings. In particular, the Arts and Crafts movement was a reaction to the industrialization of the 19th century, only to be eclipsed by Modernism in the 1930s. Did one movement completely completely squelch the other? Hardly. Out of the gyration emerged a synthesis. Bauhaus architecture, for example, the basis for Modernism, took glass, steel and concrete and gave a new aesthetic to its gritty, utlilitarian forms.
So it is, so it shall be: the Yin consumes the Yang, only to give birth to the other; the thesis gives rise to the anti-thesis within itself. Another thought to keep in mind: we shape our tools; in turn our tools shape us. And around and around it all goes.
What this tells me is that we're all swimming in the tides of history, and, like it or not, we just have to move with the flow. Contrary to the self-attentive nature of the times, it may help to let go of one's ego, to detach one's identity from one's style, one's technique, one's tools, and, ultimately, one's own handiwork. Maybe we can turn to the ancients who took inspiration from the Muses. They acknowledged that behind their art was some mysterious spirit that called on them to be collaborators, rather than I-Alone-Can-Fix-It ego maniacs. The word "vocation" has its root in a spiritual calling. I'd like think that the craftsmen who built the cathedrals of Europe, or the shrines and temples of the Far East, saw themselves engaged in a vocation. They wanted to say something through their art, something that would outlast them long after they'd all returned to the earth.
Sure, as artists there's an economic aspect to our work. From that perspective AI Art is an economic threat. Like Charlie Chaplin's character in Modern Times, we're caught in the inhuman gears of digital industrialization and automation. But there's more to art than the dollars and cents of things. Yup. And intuition tells me that there lies the door to acceptance and adaptation, the door that opens up new opportunities and experiences.
Cheers!
I am increasingly using the image2image function in Stable Diffusion rather than just typing in prompts and many of those images I am rendering in DAZ studio or photographing with my iPad camera
so no, I don't see it making either obsolete
I could paint or draw my feeder images too
I am enjoying it as a filter and I like to see how much control I can have over the results
just random really doesn't satisfy me and Midjourney was too random IMO
this still is but am hoping that will be improved with new releases
some experiments using a selfie video and the same seed, different frames and the same frame, different seeds consecutive numbers then the same frame and random seeds
all with the prompt ugly witch in front of shelves of potions
Anyone using audio input?
as Bob Ross says, art (painting) is a pursuit interest. You don't do it for anyone else or to wow the world. Just create what you want to create. Bob Ross hardly made paintings that were even close to some works of the Great Masters, but that didn't stop him from enjoying making his paintings, or saying "what the heck, I am never going to be Rembrandt or Rubens, I am going to quit."
But that's the thing. This new AI allows literally anybody to pretend to be Rembrandt or Rubens by simply typing the words "in the style of Rembrandt or Rubens".
That is the scary part of this. You don't have to know how to paint in their style. You don't even have to know who they are other than their name. You only need to be able to use a keyboard or other input to tell the AI what to paint. So the issue is that this doesn't become 'just another tool in the toolbox', it actually can become the ONLY tool that some people use at all. Maybe some light touchups with Photoshop, which can also use its own AI. So you don't even need to know how to use Photoshop that much, either.
That is where the doom and gloom is coming from. While we can say that this is like any other shift or movement in the art world, it really isn't. This is a whole different ball game from anything the art world has ever seen. Anybody can become Rembrandt. If everybody can become Rembrandt, then what makes anybody special? If everybody does something, it is not special anymore.
That doesn't mean that it is all bad. I am sure there will be a lot of cool things to come out of this. Nvidia today revealed that you will be able to make game mods with AI to give old games a big facelift. You basically upload the game into Omniverse and it can handle all of the stuff for you (though they didn't show the actual process, so that remains to be seen.) That is pretty cool. But that cool factor can be flipped. If the AI could make large portions of a game, then what happens to all the people who would normally do that? There will be fewer of them, that is for sure. They will not go extinct, no, but there will be studios that use the AI to do work that people could have done in the past.
This is already happening, BTW. Many of the remasters we are seeing today are just AI upscaled versions of the original assets. The GTA remasters were lambasted for this, as the assets contained wacky spelling errors because apparently the AI thought some letters on signs were other letters. The upcoming Suidoken remaster has many AI upscaled assets of the original game. Though I suppose it can also be argued that the low cost and ease of these upscaling techniques played a huge role in these remasters getting made at all. It is very much a double edged sword.
Like the saying goes, with great power comes great responsibility. That is where new rules and laws have to come into play, to keep this great power from being overly abused.
As I binge watched a bunch of these music videos that feed AI lyrics to make pictures (like I linked above,) I began to realize something. Here I was watching these music videos, and not the original music videos that were produced for them. Nor was I watching concert footage. By watching these videos I cut those creative people out of the process. Sure, I watched these out of curiousity, but I found a bunch of people were doing this. Just like that the flood gates had opened up for anybody to make these. There are a stunning number of people making these videos. What happens if people watch them for more than just curiousity? What if they enjoy them and want to watch them? That is where I am going with this. We have to be very careful of where we allow this to go.
And I haven't even touched on the morality of deep fake technology.
But I still believe that enough people will stand up that laws will be made to make it clear how these can be used in the future. Everything hinges on that.
Very promising. Looking forward to more.
Very promising. Looking forward to more.
Well, I somehow don't think anyone will ever mistake these for genuine Rembrandts.
It doesn't change much, ever since Marcel Duchamp placed a urinoir in an art exhibition (or at least attempted it)
https://www.bbc.com/culture/article/20170410-the-urinal-that-changed-how-we-think
Maybe it is great liberation now anybody can be an artist without the effort. The point I was trying to make earlier with Bob Ross : Everyone has a pen, but not everyone is a writer.
Everyone can pick up a pen and become a writer. Nothing is stopping you. This tool will be the same as the pen, as the urinoir, and it won't stop any "true" artists, people who want
to create, and the talents will bubble up, and time will tell if it's art or not.
I finally got some of Visions Of Chaos's Machine Learning features working
was a nightmare of installing python 3.9 with paths having to be mapped, Cmake, Git, Visual Studio, lots of checking in Command prompt
ghhahhhhh
https://softology.pro/tutorials/tensorflow/tensorflow.htm
sadly many things I want to use still throw errors
I cannot get Depthmap creation working
but I can batch remove backgrounds to create png billboards
I got the Deep Dream Video Zoom working but not Stable Diffusion Deforum
(it runs out of VRAM, Deep dream painfully slowly uses my CPU instead)
update found more settings I can lower and Deforum now works
https://youtube.com/shorts/_dXtLx4im2k
video cat zoom
https://youtube.com/shorts/4bYvjI4pY8s?feature=share
selfie furry
shorts don't embed here
All I think of when I see AI is Ain't I.
Some of the images are extremely nice though and I suppose you could goof off and generate a fairy tale's book worth of images and then make up your story around those. Just like in music sometimes the lyrics come first but far more often the music comes first.
I am finding Visions Of Chaos Deforum Stable Diffusion animations pretty awesome
except I can only do 512x512, any length, could go for hours if I wanted
any other dimensions even smaller overall crash my computer
not just my graphics card, the whole PC Blackscreens and has to be powered off and on, cannot do a hard restart
Well, I disagree. Your renders are some of my favorites and precisely what I wish I was able to do. On the other hand, nothing posted in this thread remained with me once I closed it... not that I follow it too closely, but I did check it every few pages.
The whole thing seems as exciting to me as watching a playthogut of a game on yt instead of playing it yourself.... I'll never understand the appeal of that, yet it seems many folks like it.
A big part of the fun of this is the creative process... staging, posing, lighting, tweaking the barbie's textures and dressing it up, building a story if it applies, even making some shoddy assets when I can't find what I need - if I didn't enjoy this part, I'd never spend hours on it, only to get a final image. Even if the final render ends up boring, it's still not all lost since I still have fun setting it up and often learn something new from it.
I'll say this though, it's going to be awesome for anyone just trying to make a quick buck... at least until those paying for renders will realize that instead of telling someone else what they want, they could just tell it to the AI directly for free.
That's exactly where I'm worried. I work an office job, and don't/haven't ever made money from my artwork (I don't count commissions I've spent hours and hours and hours on and only got $75 as making money)...but there are those who do make their living from artwork...whether it be book covers or artwork commissions. I worry for them simply because the people who might have paid for their time in the past might just go and hit that "Robot, make me some artwork" button and never look back.
And I really do appreciate that you like my artwork (it means a lot just to hear that).
I am a traditional artist who works with oils and watercolors. However, about 30 years ago I started using CAD, which led to me learning various 3D programs (Maya, Zbrush) and I've been using Poser/Daz for nearly as long as they've been a thing. I also do a lot of digital painting these days, using Painter, Photoshop etc. I guess what I'm trying to say is I find a lot of different ways to create Art. AI is no different. My process is no different.
My process is, let's see, say I start with an idea or three. Most of the time it's not ground breaking, but I'll think, hmm, a girl in freefall with spirit birds around her. Then I'll think about style, I like dreamy, dramatic things. Folk tales, and fairytales, and folk horror. Bohemian things, art nouveau, things that are about the beauty of things rather than function. THEN I do my composition, and when i do, I think about shapes and the big idea. What would make me feel like I'm floating in air, vulnerable, powerful? The gesture of the pose of the body? And so on. Then colors. Again, I'm drawn to drama, so contrast, vivid, or softness, something unusual. Then I think about symbolism of the narrative. Falling? Feathers? Water? And the spirits. Do I add things for interest, like sparkles, or stars, or ribbons, or bubble or clouds? How do I layer in textures, just subtle enough to not compete with the storytelling, but giving visual depth?
Well. Let's say I want to make this with an AI prompt. I have to think about all the same things, and even then, I have to play with it, add this or that, because my understanding and the AI's might not be the same. And even if I don't get what was in my head, I get dozens of thumbnails for inspiration. Plus, the entire process is focused on my aesthetics. The results tend to resonate with me, look and feel like something I would do, or want to do. The advantage is mostly time. I can find my way to my goal much faster than if I had to browse other artists, magazines, movies, books, dreams, waiting for the pieces to fall together. I can expedite this with the AI, like using a calculator for math. I can show my work and write it all out, or press some buttons and get the same answer. In the end, the result is MINE.
Sure, it's possible to just fat finger your way into it, and just 'pressing a button and getting a drawing' ist rather like something raw, like a 3 year old with a crayon, or acrylic pour painting. At its most fundamental, it's art - just not very good art. I've been poking about with Art AI since nearly the beginning. It's exploded into our collective world because it's progress is exponential and in the last 6 weeks it's become ground breaking. But have you spent time looking at the AI art being produced? Most of it is meme quality stuff, some of it is interesting, but noisy and far from professional work, and only about ten percent is what I would consider noteworthy. Just like anyone can dump paint on a canvas and call is 'abstract', but it takes someone with some artistic insight and ability to make that process Art.
You quickly find out that yes, you can type, "A bird by Rembrandt" and get something that, in the beginning, is amazing to you. After a while you realize that all of that mindless prompting is fun for the lay person, but it's a lot more work to create real art with it. I have an MFA in Art but I have learned more about artists, particularly contemporary artists, and the elements of art, in a short time, than I did in 6 years of university.
Now that SD(DreamStudio) allows image input, it's opened up yet another level. I take my own art and get to play with variations of it, and it's again, a short cut to artistic evolution. It influences my vision. I am planning on creating a graphic novel, using Daz and AI (and of course, photoshop, Clip, etc) All of that was possible before, but now? I can make it MINE in ways I couldn't before, short of hand drawing it myself.
I get it, this is all very scary. It's a game changer. Art will never be the same. Compare it to when Photography came about and changed the Art world. This will do the same. My friends in the gaming industry are already figuring out how to change their pipeline to use AI. But, people will still paint and draw, and the world will adapt and change. For there's no putting the genie back in the bottle. AI right now is so good you cannot tell if it's traditional or AI. There's no way to judge something if someone continues to use it and it becomes 'forbidden' in some corners of our world.
And if you think Art is the only thing going to get shaken up, wait until you see what's coming with Music, and Writing. Faster than you can imagine. Months, not years.
.
Anybody know if the galleries are working? I just posted (well, I think I posted) something but it seems to have utterly vanished. And to stay on-topic, yes it was an AI assisted piece, the first one I consider up to par.
EDIT: I seem to have found it! Clicky the linky & tell me what you think! https://www.daz3d.com/gallery/user/6386104575983616#gallery=newest&page=1&image=1248862
It's an interesting state of things I'd say.
The banning is on.
"A few weeks ago, some artists began discovering their artwork in the Stable Diffusion data set, and they weren't happy about it. Charlie Warzel wrote a detailed report about these reactions for The Atlantic last week. With battle lines being drawn firmly in the sand and new AI creativity tools coming out steadily, this debate will likely continue for some time to come."
https://www.theverge.com/2022/9/21/23364696/getty-images-ai-ban-generated-artwork-illustration-copyright
https://arstechnica.com/information-technology/2022/09/flooded-with-ai-generated-images-some-art-communities-ban-them-completely/
Been following that. No doubt there'll be plenty of debate, regulation and maybe even legislation. But. You cannot put the toothpaste back in the tube.
https://www.theatlantic.com/technology/archive/2022/09/dall-e-ai-art-image-generators/671550/
Not surprised at all about this...
No but you can control how the toothpaste comes out and this is what I think will happen.
No control is bad. On the other hand, too much control and/or control for thee but not for me also sucketh mightily. Finding a reasonable, workable and (hopefully) fair balance is going to be the trick.
Agreed but until the details are settled using anything ai for a commercial product would be a gamble.
And then sometimes you make something like this in seven seconds like I did this morning and wonder why you keep messing with 3D at all.
We barely even have 3D clothing as detailed as this, let alone this style being almost impossible to replicate purely in 3D. I want to keep using my 3D figures and models, but it's incredibly frustrating that no matter how much I spend and how many programs I own, they are not going to give me results like what this did in less time than it takes for Iray to even warm up - or for that matter, if I spent a year trying to do it myself.
Video AI wrote the music too (I arranged it)