Stable Diffusion, NovelAI, Machine Learning Art - AI art generation discussion and image dump

To anyone worried about this killing art, this is literally the equivalent of a fleshlight to an actual coochie. Women haven't died out yet.
The only people I'd say are in danger of losing work to AI are concept artists for big companies who will be more willing to use an AI to create their concept art for movies and games, or hire a wrangler to tell the AI what to do for much cheaper and faster than an actual artist could make. People who are using AI over commissioned art aren't people who were part of the commissioning market in the first place either because they don't care about working with and supporting artists or they couldn't afford commissions. The consumer loss is negligible. Besides, most of the people I see using AI art are doing it for fun or are brainless coomers who need to see 600 AI generated eldritch holes on anime girls in 3 seconds in order to get off.
 
Here's a bit of an advice: try adding ", horrifying" at the end of your prompt. The results are out of this world.
Also, when you do that, don't set the CFG Scale too high up - no higher than 6.
tmp67zx__ds.pngtmpk0swjuoy.pngtmpg7zvmycp.pngtmp8boptjdh.pngtmpdt2hr8gi.pngtmpynh_4vdt.png
 
My man what you're describing is the only invention older than the stick or rock. It's called an imaginary friend and you don't have to pay anybody for one.
This is advanced Tulpamancy.
 
  • Like
Reactions: Fireman Sam
Did artists disappear when the camera showed up? No.
But I don't believe it's considered transformative if you take a picture of a painting and try to sell it. There have been several cases where editing existing art slightly still means it's an infraction, even turning a photo into a statue. The problem is the law not catching up to cover this and it won't for several years. But at the same time this creates quite an open landscape of creative expression that has been stamped down by years of draconian copyright creep, so that is good.
 
To anyone worried about this killing art, this is literally the equivalent of a fleshlight to an actual coochie. Women haven't died out yet.
The most concerning thing that I've seen people have a reasonable fear of isn't of being generally replaced- even if an AI could do literally anything, it wouldn't replace artists, in my opinion. The more tangible fear is of artists being replaced by themselves. There's already been several artists who have had their work fed into algorithms to produce material similar to what they make- and utilizing their name as a tag to specifically get it. Fascinating as that is, it's pretty threatening to be at risk of being run out of your work by a mechanical duplicate of yourself, and angers people too, because their work was put through a machine to duplicate it without their consent or knowledge.

Solving that issue of random people using an artist's own work to nullify them isn't something to just snicker at I think. The one thing I think could be hugely damaging to the algorithm development scene is people getting sick of their shit getting ripped and fed to learning machines, so they take it to court and put the development company through the wringer for something along the lines of using stolen intellectual property. After all, the algorithms have internal references, and if they lose a huge chunk of them, then you're back to square one. I think that's something that AI developers have to get ready for the possibility of.

Here's a bit of an advice: try adding ", horrifying" at the end of your prompt. The results are out of this world.
1666132874290.png

Holy shit it turned Keffals into Peetz
 
as someone who has zero artistic ability, this and that thing that transformed your nonsensical doodles into really landscapes/formations, I am loving the shit out of this. I’ve spent hours generating weird hellscapes, images that would make Junji Ito reconsider his life choices, and food. These various models create some ridiculous looking hamburgers.
Do you mind if you share some of your tips/generated images? I'm pretty sure many of us are interested in generating hellscapes from doodles/words (I think I'm in the minority in that I want to use this for fantastical landscapes more than anime booba). Curious what parameters and input worked best for you and all that.

It's only 28 pages in, but I feel like this is turning into a great little mini-art community. Happy to be a part of it. :)
 
Can someone put in "Final scene in Deltarune chapter 7 before the game ends" and see what it spits out?
Seems like NAI prefers multiple panels, people with their back turned, walking away, doors, and long shots of hallways. Didn't really spot any good renditions of the characters, though, probably not enough in the training set without guiding it to a specific one. Did have a lot of purple. I like the overall atmosphere it can generate. Here's a couple.
Final scene in Deltarune chapter 7 before the game ends s-2927854707.pngFinal scene in Deltarune chapter 7 before the game ends s-434346627.pngFinal scene in Deltarune chapter 7 before the game ends s-474953197.png
 
I was given the idea of "improving" a few picrew profiles pictures. Had a hard time finding any male (male) ones on google, which really makes you think. Getting the ai to do the one thing it's bad at, drawing hands, and poking out tongues apparently.

COME ON TROONS! GET YER ANIME GIRL AVI'S TODAY!
The most generic tranny I have ever seen without the inclusion of a trans flag
1666131149839.png1666131556190.png

Racist gamer
1666131270799.png1666131806879.png

Discord kitten
1666131290879.png1666132648572.png

Mastectomy victim
1666131318649.jpeg1666133460591.png
 
Japanese art community is passing around a Google form which asks you if you can tell the difference between ai-generated art from this tool and human made art. The results may surprise you. Some prompts are NSFW.

The quiz is here.

https://docs.google.com/forms/d/e/1FAIpQLSdhEpBRnOwiFI-ieNKKu3Y0KcoFbd_ZWod1LeyoV6EEfY78HA/viewform


Results archive:

View attachment 3746640

View attachment 3746641

View attachment 3746642

View attachment 3746643

View attachment 3746644

View attachment 3746645
slides which show what people voted for example:
View attachment 3746654

In conclusion, no, ai is not replacing experienced artists at the moment. It IS replacing artists with poor skill and abstract or minimalistic style, however. Ai has poor understanding of three-dimensional space around characters that create awkward scenes and poses. Experienced artists are able to tell which are ai and which are human, but some ai work is bewilderingly good.
34/45, not bad for someone who's an artistic retard. AI really doesn't understand detailed background landscapes, and it blends scene transitions like a first year art student.

AI is most interesting to me for modders, adding low-quality fake voices to characters without having to beg for actors will make modding for social retards easy.

In the same vein, AI is the death knell for voice acting. High quality models and NN can take a sample audio input from a VA and create any number of lines.
Well said botanon.

There's a new-ish repo with easy-to-use webui and a fairly easy install guide: https://invoke-ai.github.io/InvokeAI/. By fairly easy, I mean it's going to be easy if you work in the IT/dev space, normies will need to RTFM or pay to use various online tools.
 
Seems like NAI prefers multiple panels, people with their back turned, walking away, doors, and long shots of hallways. Didn't really spot any good renditions of the characters, though, probably not enough in the training set without guiding it to a specific one. Did have a lot of purple. I like the overall atmosphere it can generate. Here's a couple.
View attachment 3748446View attachment 3748447View attachment 3748448
I knew one of the most likely outcomes was Kris causing third impact in the real world and forcing all the lightners to go to the dark world and seal them off before being killed.
 
  • Like
Reactions: Mr.Miyagi
I hate it but it's also something that everyone knew was coming. Great for memes though.

I'm also sure that ever since the industrial revolution happened everyone took a look at their news source at the time, saw the headlines that said "NEW MACHINE CAN DO X STUFF!" and thought "nah, they will never replace humans" as a cold sweat dripped down their eyeballs.

It's not a matter of "if", it's a matter of "when" , i can see artists losing visibility due to people dumping AI-generated smut into a site like, say, Danbooru or Pixiv, i can also see them losing comissioners because they can generate what they want in a faster (albeit not that refined, for now) manner. It will affect people in more ways than just "oh, it draws better than me". Imagine being a concept artist watching this shit churn out content way faster than you can.

I'm sure you can also open other cans of worms and go into the "is it art even if a machine made it?" or "artists learn from their own sources and so do machines, what's the difference?" but i feel like that discussion would be overly subjective.

That said, we are kinda already governed by a hand-rubbing AI that will happily replace a human with a machine the moment it becomes profitable to do so, so i can see what's coming down the pipeline next.
 
Last edited:
  • Like
  • Horrifying
Reactions: GoPro and Tree
Here's advice that's so simple that it's dumb: Try adding keywords like "high quality" "highly detailed" "finely detailed" "masterwork" etc. to your prompt. It'll pull in vectors containing more high quality sources. "portrait" "bust portrait" etc. also work surprisingly well for people. Also don't be shy about adding artist names which are incredibly strong vectors in SD and will change the whole mood of the picture. "oil on canvas" etc. works well too. Capitalization doesn't matter, the AI doesn't have a concept of it. This also works in the inverse for the negative prompt, which btw. has no real limit/overhead by the nature of how it works. Be careful though, as everything with AI is subtle, sometimes you might pull in/block things you didn't really mean to just because the AI made a tentative connection. Photography keywords like "zeiss" "35mm" etc. also work pretty well.

ezgif-3-f18f06c662.gif

The face is AI generated, the movement was made with DeepFaceLive. (Not my work)
 
Solving that issue of random people using an artist's own work to nullify them isn't something to just snicker at I think.
Oh no, that I actually agree with which is why I don't use these things. When an AI makes it from scratch, it's a marvel. But these things are basically photo mashing others people work to get these results. Photo mashing is a legitimate tactic but at the same time, it also feels like an insult to people who spent years to reach the abilities they have that all a bot has to do is mash several of their images to make something that's supposedly on the same quality level as theirs without much effort.

That I absolutely get, but the technology itself isn't bad. I see good usage of it for artists, just the current market is more focused on using it to make free degenerate porn sadly.
 
Back