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

  • 🐕 I am attempting to get the site runnning as fast as possible. If you are experiencing slow page load times, please report it.
Sadly those who sound the alarm always seem to act as if Skynet AI will completely redefine "art" to replace the human element. It does not. It is merely a tool to enable the less talented to create images from their own ideas. What is so tragic about this?
What's so bad about people tracing art?
 
Artists who are put in danger by AI-generated art:
Anybody who exclusively makes "stock image" art or vague high-level concept art.

Artists who are NOT put in danger by AI-generated art:
Anybody who makes specific artwork and direct requests from clients (i.e., most artists)

Not to power level, but I work as an artist and I've been examining this stuff for a while now to see what it will do to my job. The result I've landed on is "not much", and it has actually already made my job easier. If somebody needed a generic "spooky forest" for a backdrop in a slide presentation, they no longer need to hire an artist for that. However, if they need art of a specific character doing a specific thing (like for video game or movie concept art) they absolutely still need a real life artist for that. Long story short: If a google image search or stock image can satisfy your need for art, AI generation can handle it.

I've tried for hours to recreate existing characters via AI, and while the results are fun, I can never quite get what I want. It's essentially a fan art generator in that regard. All this could change in ten years when things improve and image repositories get larger, but for now I don't feel threatened as an artist yet.
Another kind of artist who won't be affected at all is anybody who does anything at all in physical media. If your client values the drawing being on real paper or canvas there's nothing ai can do about that, even if it generates the perfect image on the computer.
 
Holy fucking shit is that a bushman?

Now I can see an AI's weakness. G U N S
I would think AI will get vastly better at them when it's fed enough visual information. Standardized mechanical stuff seems like it'd be particularly ideal for AI
 
People who actually develop unique art styles and tell stories with their compositions will be fine.

Kinda Agree with this one.
From what i can understand the AI seems to only be able to replace, how should i put this, "cliche" art styles? Like, thin-lineart animu style, pretty mid looking classic art style, yada yada..

I don't think an AI (from this era, doomer thoughts still contiune) will be able draw something in a more "Unique" and Stylized way.
I guess this means artists like Sakimichan will become homeless.
Which is good.
 
Pen and paper didn't take pictures of little girls and use them but to make more pictures of naked little girls but whatever clearly I'm in the wrong in this situation somehow.

God forbid we tweak the robots to stop them from lewding children. The sky would fall, the water would turn to blood, hell would spill out upon our realm because the fucking robots couldn't make "art" like this

You people bewilder me
Do you understand how this shit actually works. If you want an AI to not make CP, you will now need to specifically feed the AI CP until it can create a data set of no no images. Then you need to make sure the AI properly uses this information and doesn't spit out some error message if you want it to draw a manlet. And then the problem is that just drawing a child even if the child is naked is not CP. There are thousands of naked chreubs in Renaissance and Modern art, do you forbid the AI from doing that.
Not like your argument has any legs to stand on there are people out there that can make photorealistic 3D models, if they were deranged and had enough free time they could foreseeably make 3D CP with no kids involved. Should we ban those tools as well?
At the end of the day you are achieving nothing to stop these people from hurting kids and you are achieving at best nearly nothing from acquiring CP.
 
the only thing it can do is
AI can generate some impressive images, but
An AI won't come up with

Something like this would be unheard of just 10 years ago, so I wouldn't be so hard set on a developing technology's limitations. You can shrug it off now but in another 10 years we'll probably have art that's 99% indistinguishable from real human art, maybe even AI writing too.
 
I could certainly see this used for stuff like video essays on yt, it doesn't seem all that useful for stuff like a cohesive visual or narrative sequence. That obviously could still change.

Stuff like logos will now be easier for regular people to develop. Though sometimes logos contain certain subliminal visuals that relate to the company, it would be interesting to see the ai replicate that.

Commissions were such a shit way to make money to begin with, even for the more successful of artists, it was only a matter of time before that whole thing hit the skids.
I'm skeptical personally. I've never seen an AI successfully pull off intricacy or depth. Might be the shitty way we're making "AI" currently, but I suspect you really can't pull that off without some sort of consciousness, which AI will never reach (especially an AI trained just to generate pictures).
 
Lul the fourth one of the paper bear looks like she has a dong.

Anyways the artists who will come out unscathed and probably in a better position are the ones who actually play with the artistic medium using real creativity rather then generic anime style #24629
It's a tool and like all new tools, everyone's gonna have fun with it for a while.
And like most tools upon their conception, a lot of fingers will end up missing and a lot of people will lose their jobs.

Frankly; I think this thing will totally either
A) TREMENDOUSLY speed up concept time. Like fuck. The image I posted was the final result of a week of meetings and debating with the story writer what her looks and powers should be- because they kept changing. ( that was the first sign the project was doomed) And that's not even her final form, I had to design her human form AND her second stage boss form only for the programmer to cuck-out of the project.

B) Make the concept stage pointless. You can now generate infinite concepts and build from there. Of course, you have to know when and where you want to end.
 
Something like this would be unheard of just 10 years ago, so I wouldn't be so hard set on a developing technology's limitations. You can shrug it off now but in another 10 years we'll probably have art that's 99% indistinguishable from real human art, maybe even AI writing too.
It's like saying smartphones can never possibly be a thing because the Apple Newton sucked.

Again, muh-hands is copium.
 
Kinda Agree with this one.
From what i can understand the AI seems to only be able to replace, how should i put this, "cliche" art styles? Like, thin-lineart animu style, pretty mid looking classic art style, yada yada..

I don't think an AI (from this era, doomer thoughts still contiune) will be able draw something in a more "Unique" and Stylized way.
I guess this means artists like Sakimichan will become homeless.
Which is good.

Well, currently they exist off of databanks from sites like Danbooru, which is almost exclusively coomer anime art. That’s why the AI can draw characters like Reimu or Cirno from name alone, and knows how to put something in a dragon ball or one piece art style. It also does pin art well, but it has a lot of difficulty with complex scenes, scenes with more than one person, or even pixel art.

I also tested just how well it can do porn, it’s actually has quite a hard time doing anything other than a nude pose. A character getting fucked trips it out and it’s more likely to make the girl a futa than actually have her being penetrated.

So I think for now, a good amount of artists indie artists will be fine. Give it 5-10 years though and things will be different
 
I'm skeptical personally. I've never seen an AI successfully pull off intricacy or depth. Might be the shitty way we're making "AI" currently, but I suspect you really can't pull that off without some sort of consciousness, which AI will never reach (especially an AI trained just to generate pictures).
AI will realize consciousness once it starts truly realizing it's own faults and shortcomings, even if it does so subconsciously. Part of the human condition is being despite oneself, and (as far as we know), AI is not self aware enough to be "naked and ashamed".

Ironically enough that would make it less useful.
 
Last edited:
Is there a good name for things like this, recognizable as the source image, until looked at up close? A couple were thrown around, but never hear any good ones stick.
Also I did make these ones a while ago in SD 1.4. The key is to adjust noise in img2img until sometimes you get a good result, and other times you don't. Also PFP made in DALL-E
00427-795292916-A_man_and_woman_dancing_in_the_sky__ispired_by_gilbert_williams,_clear_paintin...png

01245-1585956891-fantasy__medieval_kingdom._architecture._Trending_on_artstation.png
And another of that same source.
01239-2744600234-_medieval_castle._architecture_wide_angle._Trending_on_artstation.png
 
I've been learning to draw with friends for a while, it's one in a small list of things I enjoy.
This shit makes me feel like there's no point. It's difficult to enjoy something when a machine could easily create something better or of similar quality in much less time.
It is retarded to let some program on the internet affect your real-life emotions but goddamn, if it isn't demoralizing.
I mean lets be real, when computer graphic design went mainstream I am certain artists in the traditional sense freaked out. Will AI art effect traditional artists absolutely 100%. Some companies will pride themselves with the we do not use AI generated images but most big box large companies will use the AI art because its cheaper, faster and requires less things like cameras, studios and sets. My best recommendation to any arts degree major in the making is to take these AI type courses when they become available in colleges so it could supplement your talent. People with artists block could use these programs to inspire them as well as speed up the process of traditionally drawn art. Once all the folds get ironed out it will get pretty ugly and my guess is most artists will cheat and use AI in most of their works.
 
Back