Shit post, not shitpost
this post was submitted on 05 Feb 2025
92 points (84.8% liked)
Lemmy Shitpost
27928 readers
4026 users here now
Welcome to Lemmy Shitpost. Here you can shitpost to your hearts content.
Anything and everything goes. Memes, Jokes, Vents and Banter. Though we still have to comply with lemmy.world instance rules. So behave!
Rules:
1. Be Respectful
Refrain from using harmful language pertaining to a protected characteristic: e.g. race, gender, sexuality, disability or religion.
Refrain from being argumentative when responding or commenting to posts/replies. Personal attacks are not welcome here.
...
2. No Illegal Content
Content that violates the law. Any post/comment found to be in breach of common law will be removed and given to the authorities if required.
That means:
-No promoting violence/threats against any individuals
-No CSA content or Revenge Porn
-No sharing private/personal information (Doxxing)
...
3. No Spam
Posting the same post, no matter the intent is against the rules.
-If you have posted content, please refrain from re-posting said content within this community.
-Do not spam posts with intent to harass, annoy, bully, advertise, scam or harm this community.
-No posting Scams/Advertisements/Phishing Links/IP Grabbers
-No Bots, Bots will be banned from the community.
...
4. No Porn/Explicit
Content
-Do not post explicit content. Lemmy.World is not the instance for NSFW content.
-Do not post Gore or Shock Content.
...
5. No Enciting Harassment,
Brigading, Doxxing or Witch Hunts
-Do not Brigade other Communities
-No calls to action against other communities/users within Lemmy or outside of Lemmy.
-No Witch Hunts against users/communities.
-No content that harasses members within or outside of the community.
...
6. NSFW should be behind NSFW tags.
-Content that is NSFW should be behind NSFW tags.
-Content that might be distressing should be kept behind NSFW tags.
...
If you see content that is a breach of the rules, please flag and report the comment and a moderator will take action where they can.
Also check out:
Partnered Communities:
1.Memes
10.LinuxMemes (Linux themed memes)
Reach out to
All communities included on the sidebar are to be made in compliance with the instance rules. Striker
founded 2 years ago
MODERATORS
I disagree. I also think your anger should be redirected at the AI companies stealing all the art rather than the people having fun in a sub dedicated for AI.
I'm not particularly fond of AI art, but this is a pretty petty point to make when companies are scraping everything from everywhere
Good point, I know I was just trying to poke fun at "making" AI art at first but I might have gotten a little carried away!
aye, fair enough!
Well... a person did make the art. Several in fact. ELIZA just took what they made and put it in a blender.
He didn't make poetry he only put the WORDS ON THE PAPER? THE Pencil WROTE IT!!
TRUE ART IS CHISELED!!!!
-Pedantic jealous non-artists like you lol
Never said I was an artist. Also if the pencil wrote by itself you might have an argument.
The anti ai conservatives are a strange group. Go be mad about something that matters.
When did I say I was anti-ai or conservative. I just didn't like my comment being removed when I said making art involved more than typing in a prompt or 10 and hitting enter.
At its core, conservatism is about "conserving the way things are", or, in another way, being "against change". Anti ai people are against the changes they're seeing. I don't know how else to describe this.
Art is art, if someone has a fun idea and uses ai as the tool to express their idea, then it's art. It may require less effort at the time of creation than other art forms, but this was at the expense of many people spending many hours developing the tools to make that happen.
Just like every era of change, there are people out there pissed off because it's different. People were upset that the printing press was created, because they felt it cheapened knowledge.
It doesn't even make sense. Conservatives love, love, love AI.
Hey, does anyone remember that 500 billion dollar infrastructure package Trump wants to give AI companies? You should, it was two weeks ago.
It's more complicated then that, convervatives generally want to control anything that is new and gives them power, owning the most advanced AI developments is a pro to them, regardless of if they like it or not.
It's also a bit of a handout of course to Trumps buddies who got him elected in some fashion.
Prompted it is the best one I've heard, as crafting the right prompt is something of a skill on its own.
They didn't make it, they prompted the AI to make it.
This is how art has worked for millennia.
I go to a human artist and say "please make me a painting of my family. Make my wife more beautiful, me more tall, and my kids not look like little shits." And then you give money. That's a prompt for a commissioned work.
No one ever praises the Duke of Milan for commissioning a painting of The Last Supper. They praise Leonardo Friggin' di Vinci for making it.
Yeah, that's a great analogy actually.
It is not. The issue here is not "praise".
It's weird to see this used as a con. This is exactly the framing AI app creators use. If it's the same as paying an artist to make you art, then there are no issues whatsoever with using that art wherever.
There are deeper questions here.
This is weird to have to say, because I feel like I'm more open to AI generation than most online people, but I still think it needs a new copyright framework, it's not the same as buying work from a human.
Honestly, these are both simplistic takes. Tools are tools are tools, so for better and worse there is a big gradient across "popped up a genAI and asked for an image with a prompt">"asked a genAI for an image and then asked for changes">"asked a gen AI for an image and then fine tuned it and tweaked it in parts to the point where, let's be honest, this would have been faster to do in Photoshop from scratch" (which is a thing and it baffles me a little)>"made an image using some genAI content in the process".
It's a bit of a mess, and a good reason why to be safe for production and commercial use a lot of pro content creation places just ban the technology outright to avoid legal issues later. That's a good call.
If you want to draw a line for amateur image sharing... I don't know, it gets weird and complicated.
Well it's like you commission a work when you use AI. Do this, more like that, bluer, smaller etc. So not creating art IMO.
there's also sometimes deeper collaboration with image generation, where you edit the result in like krita and then send it through the generator again to smooth out the edges of the touch-ups.
i occasionally do work where i cut up existing images to make new ones, like making a person in a particular pose by pulling an head from one image, a torso from another, an arm and background features from a third etc, then i would painstakingly color match and blend it all together with hours of filter, clone brush and single-pixel edit work. with image generation i can get a blending and lighting pass for my image in a few seconds. it's not perfect but the speed makes the work a lot easier.
load more comments
(3 replies)
load more comments
(3 replies)
You're 100% right. Hell, I've been there. I imagine your comment got deleted because the point of the thread wasnt to debate about the definition of artistry.
People who ask for art didn't do the art. Wretch, you have merely stated a request. Worse, you have spoken your desires to a demon, and now you proudly display its gift as your own work.
This!
It's just a commission.
It's just disappointing how few people are literate with standard occult practices. Never summon anything more powerful than you, never tell your innermost thoughts and desires to a demon, and if you are that stupid don't brag about it. Real JV league demonology.
- So really it should be "[work description] by [Dali-2], commissioned by [Name]"
Commissioned or better prompted IMO.
- It should not be. But yes.
Cool post but how is this a shitpost
How is it NOT?
It takes less effort to post a single paragraph of ranting than it does to edit the text on a meme image. In a way, this is a shittier shitpost than most.
👍
Cause its not a very important post and its mostly just me bitching lol. Almost posted to one of the conversation subs but figured it fit here as well.
load more comments
(2 replies)
I remember you. It was my thread you commented in. You're mad because a moderator removed your comment after you showed up and started antagonizing no one in particular?
I'd like to ask you a question. How much experience do you have with any Stable Diffusion tools?
I mean I've used em. Still don't think the guy you were talking about "made" a thing. I was just mad the moderator removed "AI Art Hate" when all I was saying that is basically like if you went to Subway ordered a sandwich then brought it home and claimed you made yourself sandwich.
This seems like a good place for discussion so if you'll humor me, I'd like to explain some things you might find in a prompt, maybe some things you weren't aware you could do. Web services don't allow for a lot of freedom to keep users from generating things outside their terms of use, but with open source tools you can get a lot more involved.
Take a look at these generation parameters:
sarasf, 1girl, solo, robe, long sleeves, white footwear, smile, wide sleeves, closed mouth, blush, looking at viewer, sitting, tree stump, forest, tree, sky, traditional media, 1990s \(style\), <lora:sarasf_V2-10:0.7>
Negative prompt: (worst quality, low quality:1.4), FastNegativeV2
Steps: 21, VAE: kl-f8-anime2.ckpt, Size: 512x768, Seed: 2303584416, Model: Based64mix-V3-Pruned, Version: v1.6.0, Sampler: DPM++ 2M Karras, VAE hash: df3c506e51, CFG scale: 6, Clip skip: 2, Model hash: 98a1428d4c, Hires steps: 16, "sarasf_V2-10: 1ca692d73fb1", Hires upscale: 2, Hires upscaler: 4x_foolhardy_Remacri, "FastNegativeV2: a7465e7cc2a2",
ADetailer model: face_yolov8n.pt, ADetailer version: 23.11.1, Denoising strength: 0.38, ADetailer mask blur: 4, ADetailer model 2nd: Eyes.pt, ADetailer confidence: 0.3, ADetailer dilate erode: 4, ADetailer mask blur 2nd: 4, ADetailer confidence 2nd: 0.3, ADetailer inpaint padding: 32, ADetailer dilate erode 2nd: 4, ADetailer denoising strength: 0.42, ADetailer inpaint only masked: True, ADetailer inpaint padding 2nd: 32, ADetailer denoising strength 2nd: 0.43, ADetailer inpaint only masked 2nd: True
To break down a bit of what's going on here, I'd like to explain some of the elements found here. sarasf is the token for the LoRA of the character in this image, and <lora:sarasf_V2-10:0.7> is the character LoRA for Sarah from Shining Force II. LoRA are like supplementary models you use on top of a base model to capture a style or concept, like a patch. Some LoRA don't have activation tokens, and some with them can be used without their token to get different results.
The 0.7 in <lora:sarasf_V2-10:0.7> refers to the strength at which the weights from the LoRA are applied to the output. Lowering the number causes the concept to manifest weaker in the output. You can blend styles and concepts this way with just the base model or multiple LoRA at the same time at different strengths. You can even take a monochrome LoRA and take the weight into the negative to get some crazy colors.
The Negative Prompt is where you include things you don't want in your image. (worst quality, low quality:1.4), here have their attention set to 1.4, attention is sort of like weight, but for tokens. LoRA bring their own weights to add onto the model, whereas attention on tokens works completely inside the weights they're given. In this negative prompt FastNegativeV2 is an embedding known as a Textual Inversion. It's sort of like a crystallized collection of tokens that tell the model something precise you want without having to enter the tokens yourself or mess around with the attention manually. Embeddings you put in the negative prompt are known as Negative Embeddings.
In the next part, Steps stands for how many steps you want the model to take to solve the starting noise into an image. More steps take longer. VAE is the name of the Variational Autoencoder used in this generation. The VAE is responsible for working with the weights to make each image unique. A mismatch of VAE and model can yield blurry and desaturated images, so some models opt to have their VAE baked in, Size are the dimensions in pixels the image will be generated at. Seed is the number representation of the starting noise for the image. You need this to be able to reproduce a specific image.
Model is the name of the model used, and Sampler is the name of the algorithm that solves the noise into an image. There are a few different samplers, also known as schedulers, each with their own trade-offs for speed, quality, and memory usage. CFG is basically how close you want the model to follow your prompt. Some models can't handle high CFG values and flip out, giving over-exposed or nonsense output. Hires steps represents the amount of steps you want to take on the second pass to upscale the output. This is necessary to get higher resolution images without visual artifacts. Hires upscaler is the name of the model that was used during the upscaling step, and again there are a ton of those with their own trade-offs and use cases.
After ADetailer are the parameters for Adetailer, an extension that does a post-process pass to fix things like broken anatomy, faces, and hands. We'll just leave it at that because I don't feel like explaining all the different settings found there.
I could continue if you want to hear more.
Thanks for this comment, this actually made me want to play with one where I had kind of just waved it off before. Any recommendations for one that is very light weight and can be used offline. Wait times are fine but those configurations I'm sure I'll need to find a reference manual for.
Counter argument for the sake of education, go try to recreate his picture using Stable Diffusion. Let me know how easy it is.
I don't really get how this is a counter point. I don't think anyone is contending that the pictures produced are reproducible by the same means. They're contending that the method of production isn't "making" art and they aren't an artist for starting the production process.
It's sort of like when rich people go to space and call themselves an astronaut. People have an idea of what an astronaut does and it isn't just "space tourist." If you fired back with "you try spending that much money and see how easy it is" then that wouldn't answer the point of why people don't want to call space tourists "astronauts."
I don't think their point was just that it's impossible to reproduce, more that there is skill, knowledge and choice put into getting close to the intended idea when working with AI output.
With that I think your point breaks down when you compare it with something like photography. Often you aren't 'making' the images that you capture, but there is skill and artistry in the choices that capture the moment or picture you want. Obviously there is more control in photography, and I would disagree with anyone that uses AI and claims the same level of artistry of photography. But ultimately I think the lines around art are so blurry in general, it seems incorrect to me to do decidedly exclude AI generated images.
I don’t think their point was just that it’s impossible to reproduce, more that there is skill, knowledge and choice put into getting close to the intended idea when working with AI output.
That's interesting cuz I took their point as "you can put the exact same prompt into the stable diffusion and not get the same image each time, thus good luck trying to recreate the picture." Which seemed to me to suggest the opposite point: That intentionality has a diminished role in creating ai images, so it serves even less of a role as art. You wouldn't say someone sitting in front of a slot machine "intended" to get a cherry, bell, and bar on a specific pull, after all.
Often you aren’t ‘making’ the images that you capture,
But... you are though. Images would not exist without the photographer choosing to make them. Not to mention that many forms of photography (albiet older forms) have very real physical elements to them like dodge, burn, and film development. Even without those elements though, those images would not exist without the effort, intention, and presence of the photographer. The photographer also makes the conscious decision about what photos not to take, because they don't align to their message. Intention is at every step of the process and that invites us to explore the meaning of their work.
Contrast that with AI art. The only intention you have is your prompt and choice of model. I would argue the fact that ai prompters need to "get close to" what they want their piece to say, rather than making the piece say what they want it to say, shows how starved for meaning the products are.
but there is skill and artistry in the choices that capture the moment or picture you want.
I don't disagree with what you're saying. But I will say that skill is not what makes art art. Skill can make you a better artist, but someone without skills can make art.
you can put the exact same prompt into the stable diffusion and not get the same image each time
My understanding is that if you have the exact same model, prompt, and everything else, if you use the same seed, you will get the exact same image output. That and that they called it a counter point, I took it to mean they were talking about the skill in the layers of different tools folks will use to get what they are hoping for. Because, a little to your further point, there are folks that get very into it and have like 30 different fine tuned loras or whatever to finesse every finer point of the image.
Images would not exist without the photographer choosing to make them.
At the risk of getting a little philosophically wanky about it, I guess I would argue the same for AI images. Like what exists is a nebulous connections of weights and nodes trained off stolen art that only connect in certain ways because of a given seed and prompt. Does a hypothetical random image of a muppets version of Kermit the frog as Darth Vader 'exist' without the high, half baked, prompt from someone using a free trial of midjourney?
Even without those elements though, those images would not exist without the effort, intention, and presence of the photographer.
I completely agree, and I think my point more was that with photography, you don't "make" the landscape or architecture you're capturing, but you make the image in the end.
Contrast that with AI art. The only intention you have is your prompt and choice of model. I would argue the fact that ai prompters need to “get close to” what they want their piece to say, rather than making the piece say what they want it to say, shows how starved for meaning the products are.
I will say, I respect photography more, and so much of what AI generates is soulless slop. I think that ultimately my push back is on the folks that argue that it can't be art. And just as there is intentionality in choosing what photos you don't take for photography, there is non-zero intentionality in generating 30 loose candidates, 50 fine tuned candidates, and 3 final images with stable diffusion.
Sorry if this comment is scattered, I've restarted my reply like three times trying to sort out what I even think lol.
I think that ultimately my push back is on the folks that argue that it can't be art.
I'm not really jumping in on this discussion, but I did want to add one thing:
I can believe two things at once.
AI generated media can't be art ... because the whole purpose of a generative AI machine is to alleviate the burden of decision making. The fewer places you let something decide for you, the more "art" you can imbue into your project. Art is a communicative effort.
Artists can use AI generated media ... but the points of interest, the meaning, would not (necessarily) be the decisions the machine made.
An example above, I forget if it was you or someone else, shows a pen sketch of a scene then filled in by the generator, and I think the artist there can be given credit for the perspective, the framing of the subject, the mech-suit, the sci-fi aesthetic; but I wouldn't credit them with the tally marks on her left shoulder, or the shape details of her eyes, or the various light-up displays that dot the walls.
There's also something to be said for choosing as opposed to creating outright, but I think we're losing ourselves in myopic details at this point.
The bottom line is that, aside of any ethics issues, I'm not that upset about AI media that's honest about what it is. I watch youtube channels that depend on AI for their performance art. But, AI proponents love selling this technology as a replacement for people, which is a sentiment I find... disgusting. Inhuman.
And, I find it really sad the way a person who spent the better part of their life perfecting a style and technique can be essentially shoved out of their own niche by the 10,000 style-copy images a generator can make in an afternoon. This isn't like photography, where painters and camera-snappers can coexist in separate styles of image production: AI generators can replace both.
Sorry, I thought all that was going to be just two paragraphs.
Sorry, I thought all that was going to be just two paragraphs.
I know the feeling lol
Yeah I largely agree, the fact that it is replacing humans is rough and really speaks to such a depressing view of art in our society. Even before AI, the commercialization of art was hugely detrimental, and really sucks the soul out of much of what it touches. The fact that AI attacks the starting footholds for artists to find any money at all in the industry is particularly upsetting.
All that being saying, I guess I would still push back a little bit on the idea that it can't be art when generated in it's entirety. I would argue that while yes it does have less intentionality, even at the most base level of a prompt and picking your favorite generation, is enough to qualify it as art. I may be a bit of an absolutist in this regard, but I justify it by pointing to things like splatter or fluid acrylic painting. Where the exact lines of intentionality is hard to draw. All of that to say I don't think there is much value in drawing a strict line that excludes AI work from art just because the discrepancy between choice of the artist and the choices on show is so vast.
but I justify it by pointing to things like splatter or fluid acrylic painting.
The counter I would give here is that those are just techniques. The challenge, then, is whether the generation machine can be made to do things that are interesting and meaningful. I know it can produce spectacle, but spectacle and meaning are different concepts.
I don't know much Pollock, but isn't he valued largely for his process and expressionism? I'm not making this accusation of you, you do seem to actually care, but a lot of people who bring him up seem to think that his work actually is random and unintelligible—I don't think that it is.
I will concede that the process of interacting with the generation machine to produce something is a creative one, I just don't think it's anywhere near what a lot of proponents claim it be.
I've used Suno, and my lasting impression of it is that it was fun, sometimes really funny, and overall kind of soul sucking. As a musician, there were essentially no times that I felt anything produced there was mine. It was just novelty. Some of it sounded really cool, but none of it was an expression of me or what I was really looking for.
view more: next ›