AI art is antithetical to art. Art requires artistic intent.
It could have some limited application for very early exploration in commercial art, or perhaps as very limited tools used in existing art software, but generative art is inherently pointless and you need artists to be able to do incremental iterations properly, which is required for real work, which isn’t supported yet. I’ll sure it’ll get better and more convincing, but it’s still inherently pointless to use AI for art, since the is supposed to be human expression.
I strongly disagree. 99% of the work is being done by an algorithm. It’s like if we had autonomous driving and you said you were actively driving all day, because you told the car where to go, and then took a nap in the car until you had arrived.
Photography is widely recognized as an art form, even though the scene exists independently and the photographer “simply” frames and captures the shot.
A better driving analogy might be Tesla’s current level of self-driving, where you have to keep a hand on the wheel and eyes on the road the whole time, and remain in charge of all the critical decisions. When someone arrives in a Tesla and says, “I drove here,” no one goes “ackchyually…” Even if we follow your analogy, it’s the individual’s idea to reach that destination- often a novel place no one has even been to before.
Creative individuals curate unique datasets, which can take countless hours of manual work, to create LoRAs. They often draw from their own photographs, drawings, paintings, etc., and then coordinate prompts and parameters to blend their custom LoRAs with other creators’ LoRAs/models/checkpoints to craft something unique. These creations exist only because they had the vision and put in the effort to realize it. The process can be even more involved with tools like ControlNet, where artists might even sketch an outline of the scene by hand.
A quick selfie might not be considered art, but intentional expression through creatively capturing a scene is (photography). Similarly, a quick generation via Copilot for a meme might not qualify as art, but intentional expression through creative generation certainly does.
I disagree with your analogy, as I find it overstates the active involvement of the driver (prompter) during the drive (actual image generation).
Preparation is it’s own process, whether you’re curating art you made yourself/stole from non-consensual artists, or have been finding references as an artist. Different skillset. They help the process of making the final image, but they are not a direct part of that process.
And let’s not kid ourselves about theses datasets. There’s no accountability so there’s no way to ensure that any dataset you’re getting from other people aren’t comprised of, at least partially, stolen art.
ControlNet let’s you add visuals to your prompt for greater control, but you’re still generating the image externally, and leaving the vast majority of the decision making to the model you’re using. Even if someone is happy with the result they get from a generative model and find it visually pleasant, that doesn’t make it art. The model is doing the work and the model cannot have artistic intent, so it cannot make art. It can make images and people can enjoy those, but those images aren’t something new.
They are amalgamations of most basic common denominator of existing things. It is much more like a really advanced collage that is great at hiding the seams.
When humans make art, they are constantly making decisions. Decisions, decisions, decisions. With every stroke of the pen, with every color (not just a generic pink, blue or yellow, but specific tones and shades of those), with every everything they to while making that piece, they are making a lot of micro-decisions. Those decisions are made in respect to the person that is making the art, as their personal life experiences are what dictate how they make such decisions, even if they don’t notice it.
AI art is not like that. With AI, you type a prompt and outcomes an image. The user does not have a say in any of the micro-decisions that when into making that piece. The AI it self isn’t making any decisions either, it is just making the mathematical weighted average of what images with a description with similar tokens look like, and simply copying said decisions. The AI does not decide, it simply regurgitates previous decisions.
You are looking at the lowest form of AI art and then declaring that to be all there is. There is more than just pressing generate to a prompt.
If you’re actually interested, I wrote a length post about it: https://lemmy.world/comment/12743504
AI art is human expression in the same way that the Gaussian blur tool is. It’s a bunch of math spitting out a pattern based on specific inputs.
All while currently being as ethical as the fast fashion industry producing scam versions of high fashion products.
It has the potential to be very useful in certain applications, but right now, all it really does is create Content to be consumed. Kinda like elevator music or that horrible Corporate Memphis style that has invaded every piece of corporate media/advertising in recent years. Soulless and without meaning. It’s pretty high quality slop, all things considered, but slop nonetheless.
I love Ai defenders who are ready to tell you what art is and what artists wants. Like maybe instead of recomending this cloud based bullshit app, first try to pick up a pencil actually?
I understand the excitement, but it is very much a situation of a layman trying to describe to experts what the expert and all their peers need.
I think it is just because AI has been hyped so much, and has genuinely made such impressive progress that people get swept up by the excitement, and idea that they could make their ideas into something tangible. They just don’t know the amount of consideration that goes into translating that.
The majority of people using AI will not insist you use it, aside from just trying to get others a realistic look on what the technology actually is. Just like a photographer won’t preach to a painter that they should pick up a camera. But that does not mean there isn’t benefit to derive from understanding how the other produces art. And if the painter thinks the camera is doing all the work and the photographer is a fraud, it’s probably good for them to get some exposure and realize it’s not just pressing a button. Like I explain here how that metaphor works for AI.
Most people I know that use AI are *shocker*, artists from before AI was a thing. They know how to draw with pencils, brushes, sponges, but also painting programs like Photoshop, sculpting programs, modeling programs, surface painting programs, shader production, algorithmic art. They are through and through artists. Them adding AI to their toolbox does not change that.
Right now it is not a tool. Right now it is an attempt at replacing artists.
It could be implemented in existing softwares in parts to make it a useful tool. Like a tool that could easily recolour parts of a fully rendered illustration, while still respecting the artistic intent with the form and lightning.
But right now it just spits out the blandest stuff, based on what it has identified as the most common denominators in art.
People that say that AI could be used as a tool to help artists clearly as never pickup a pencil to draw. The thing that makes an artists voice, that makes that art theirs are the decisions they make while making their art.
When you are drawing something, you are constantly making small micro-decisions with every stroke of your pencil, and those decision and how you make them is what makes art so beautiful, as no two artists make those decision the same way and each artist as a certain consistency in those decisions that evolves with them as a person. As such, art is so much more than a pretty picture, it is a reflection of the person who made it. Those decisions are also the fun part of making art.
AI art doesn’t let you make any decisions: you type the prompt and out comes an image. An image made of an weighted average of human made images with a similar description. You have no say in the micro-decision the machine makes, you have no say on where exactly the pencil strokes go. Therefore this machine is useless for artists. You might say “Just edit the image!”, but that doesn’t help either, as editing the image still doesn’t give you that micro-level of decision making. Also, editing a flat image with just one layer is just as useful as any other image form any search engine image search result. Unlike text, which can be easily edited to be exactly what you want.
I know their might be some wait to integrate machine learning into art, but right now the tools available don’t do that.
Hi, person here who’s drawn extensively with pencil as a kid but then slowed down. AI has reignited my passion for art, because unlike pencil or even digital drawing, you can iterate much more quickly, which allows me to do so while working a demanding job and having a life besides it.
You are severely underestimating the amount of micro-decisions that go into AI art, and more importantly AI assisted art. If you’ll allow me, let me explain.
Lets break it down into levels of effort and creative input, I like to refer back to photography since it has some comparison:
Empty canvas prompting.
spoiler
To me that is essentially the equivalent of taking a selfie or a random shot. On the scale of effort this is none to very little. But a prompt can be unique if you put an extreme amount of effort into specifying the exact details, just like a random shot can in fact be a very informed random spot.
You can put a rather massive amount of tokens into your prompt that all further specify this. At this point you can reasonably say you imagined at least the general look of the image before it was created. But you can’t say you had any part in actually drawing it or significantly determined how it was drawn. It’s basically impossible to get any kind of copyright protection over this unless you can back up your prompt very well, and only then would you get protection over your prompt, not what the AI drew.
Image to image
spoiler
You can feed an image into the AI, you add noise to the image, and let the AI try to remove that noise (After all, this is the exact same as it does on an empty canvas, but that is completely random noise). This means that a large part of your original image specifies how the AI will further try to denoise the image. As such, you are guiding a large part of how the AI moves forward. No other artist would likely use the same input image as you, so human decision making plays a bigger part here.
To me this is the equivalent of choosing a location you want to take a picture, and then scouting several locations to see which one works best. That’s the micro-decisions sneaking in. You are giving the AI an existing image, either created by yourself, or from previous iterations.
At this point, you are essentially evolving an image. You are selecting attributes and design choices in the image you want to enhance and amplify. These are decisions the artists makes based on their view of how the final image will look like. Every iteration adds more decisions that no other artist would take the same way.
Collaboration.
spoiler
The point where AI starts becoming a tool. You don’t start with point 1, or at least only use point 1 for brainstorming. You imagine the image beforehand, just like you would do with pencil. You can develop the image as much as you would like before going to the AI. You are making the exact micro decisions you are with drawing by hand, since it’s essentially the same up to this point. A photographer at this point would work out every fine detail before snapping the picture.
Except for the fact that you know you are going to be using an AI, so certain aspects need more or less refinement to properly be enhanced by the use of AI. Just like you don’t start the same if you’re going to make a painting, or a silhouette, or any other kind of technique. At some point, you return the image to the AI and mostly perform step #2, perhaps returning to brainstorming with step #1 if you want to add or remove from your existing design.
AI truly as a tool.
spoiler
Now to make something actually with #3, you start doing this process in iterations. Constantly going back and forth between photoshop and the AI, sometimes you spend days in photoshop, other times you spend days refining a part of the image with AI. There are also additional techniques like ControlNet, LoRAs, different models, different services, that can drastically enhance how well you get to what you want. A photographer at this point would take as many shots as they would need using their creatively controlled setup, and find the best on among them.
Sadly, most people that talk confidently about how much they hate AI just know point #1 and maybe point #2. But I see point #3 and #4. And when I talk to artists that haven’t yet picked up on AI, but if they are aware (or made aware) of #3 and #4, suddenly their perspective also changes in regards to AI. But the hostility and the blind anger makes it quite hard to get through to people that not all art with AI is made equally. We should encourage people that want to use AI to reach the point of #3 and #4 so that their creative input is significant to the point it’s actually something produced by their human creativity. And this is also where an existing artist switching to using AI will most likely start off from.
Also, in terms of time. #1 might take seconds, #2 might take minutes to hours, but #3 and #4 can take days or weeks. Still not as long as drawing those same pieces by hand, but sometimes I wish it was as easy as people would make it out to be.
I’m glad you’ve found value in it. I’ve played around with a similar workflow you describe in step 3 and 4, but I just find that it always produces the blandest version. Sometimes you get a surprising iteration, but I think being used to seeing visual patterns makes it much more obvious just how predictable the image details get.
It just ends up looking like a compilation of techniques.
I’m sure it won’t bring the same results for everyone, and that’s fine. I’m glad it’s that way and not just so easy it would consume all other ways of making things, as it would cease to be a tool at that point and just become the entire process. I really just try to make what I am already imagining since before I even put anything to paper, and so I know what I want the AI to help me with ahead of time. I get surprisingly close to the idea in my mind because of that. Much closer than I reasonably could get in other ways with the finite time I have.
That sounds like a good tool for you then.
I do art as my day to day, and there are definitely aspects of my work that I would love to have a bit of AI injection to help speed the process up, but that is much more as a tool directly integrated into the softwares I already use, like a beefed up content aware transform tool that allows me to move parts of a finished image around just a little bit, and having the AI fill in the small gaps that creates.
I see so many small ways to make the art process less painful or introduce more non-destructive editing tools, if only the AI was built into the software and actively training on the art you were doing, as you did it, rather than having it take over whole parts of the process as it is currently used.
Yes, I want it to go that way too. But that does take time. Which is why I think it’s a shame some people are really hostile to it indiscriminately. I get that the big corporations are absolutely wielding the worst aspects of it, and I hate that too. But once you go to grassroots and small corporations, suddenly there is a whole different view of it. AI is a product of humanity’s collective efforts, and as such all of humanity should be it’s benefactor, not a small group of people that just happened to have the most money and influence at the time. Especially if they had nothing to do with actually creating the technology behind it. And by making the story about them rather than the people using it as it should, it will inevitably lead to people falling behind to corporations, and widening the gap even further. Making artists even less valued than they already are. Which is something I’ve always fought against even before AI.
I mean, the hostility is entirely understandable. The current form of generative art is meant to replace artists.
It is part of what is currently devastating peoples livelihoods, although I think some companies and clients are already learning that it currently leads to lower overall quality, due to how much harder it is to implement changes based on feedback.
It lowers the overall quality bar, although it does have the potential to raise the floor a little.
The larger models that are causing this hype are quite literally trained on the work of unwilling artists.
It is the most disrespectful and clearly ethically wrong basis to build it on, and it really begs the question of whether the ends justify the means.
Beyond that, art is just not an area where we need AI. It largely hurts artists, is super energy demanding so it actively hurts the environment for no real benefit.
The energy would be so much better used solving actual problems, so more people could spend time doing things they enjoy. If some people enjoy AI generation, then that’s fine but I think it shouldn’t replace a passionate, skill-based workforce.
People who say that AI can’t be used as a tool to help artists clearly have never tried using AI as a tool. Everything you’ve written here is untrue.
Artists can manually curate unique datasets to create LoRAs. They can draw from their own photographs, drawings, paintings, etc., and then coordinate prompts and parameters to blend their custom LoRAs with other creators’ LoRAs/models/checkpoints to craft something unique. The process can be even more involved with tools like ControlNet, where artists can sketch an outline of the scene by hand. I.e., you can have precise control over where the pencil strokes go.
You’re wrong. You can’t have the same level of control as the person above describes.
Even if you train the model on your own work, it will still be the one generating every “stroke” of the pencil. If trained for it, it will do it based on how you must often do so, but you can’t clearly control it. You can’t control granular details of the process of creating the image. It’s all broad strokes.
I don’t know what your level of experience with art is, but so much of what makes art is tied up in the process of having to think through every little addition you make to an image. And by little addition I don’t mean “let’s add a person here” but “let’s do these 200 individual strokes that make up that person”. The involvement in the process is the point, and when an image is generated for you, you remove so much of the involvement and granular decision making, that the actual point is lost.
It’s like cooking with premade, pre-prepared ingredients. You can pick the dish and put it together with the stuff you buy, but you can’t control the whole process, because you’ve given up that for the sake of speed and convenience, and the dish will be different for it.
With ControlNet, you can control that very finely though. It partially combines computer graphics into the mix which is definitely not easy to get into.
But it’s different from drawing, that is for sure. Different decisions with different outcomes and different possibilities. But you see similar differences with every form of art. You don’t make 200 strokes for a photograph, or for a cut out collage, or algorithmic art, or sculpting. But you do make different decisions that are similar in nature. They shape the end product, and in a way no other person would do exactly the same. You still have to be involved in every step of the process, even if some steps are no longer done by yourself, they are replaced by other decisions.
Again, this is assuming they aren’t just clicking generate and calling it good to go, but if someone is using ControlNet they should be well above that, since the quality of blank generations is often not the best and demands refinement to anyone with an creative outlook.
This is why we shouldn’t confuse AI art, AI assisted art, and other forms of visual art, even if they all end up making an image as the end result. Something can be impressive when drawn, but mediocre if made with AI. Just like painting a scene isn’t the same as taking a photograph of that same scene, even if they end up being visually similar. Everything exists in context.
And yes, you are giving up some creative control for sake of convenience. But the question is how much. A painter that hand crafts their brushes will make a different painting had they used a pre-made brush. But we can agree I think that the creative control they lost by doing so is negligible. Artists generally lean on what they have produced before as references too. 3D sculptors can start with a cube too, but if they’re going to be making a person they will start with a mesh of a human figure, male or female depending on what they want. There is no shame in taking shortcuts in steps where it’s possible, even if it’s commendable if you don’t.
Art takes long and is expensive to produce, it would be unsustainable in this modern day to do everything from scratch every time. And as long as you focus on the parts where your decisions make the artistic intent happen, you can still make something unique and valuable in it’s own right.
That’s a cool visualisation of what kind of visual input you can feed into the process with ControlNet.
And it really makes it clear that what AI images is good for if communicating a general idea.
I think comparing AI generated or Assisted images or videos to photography is probably the closest analogous medium we have, but I think AI images are stort if in-between that and more classical art. You have more control over the more technical aspects of the image, as you can alter those things with big strokes, but you’ve given up too much control to really infused it with artistic intent.
Even when photography, where you are generally limited by reality, you can better infused artistic intent into the picture, because you carefully examine what makes that object of the picture unique.
Even if you try to direct AI models, it limit their scope they will always add whether the most average expression of what they’re adding, because that what it looks for in the training; the commonalities/averages of whatever it was trained on.
Even ControlNet is just a way to claw back a little more control over the process. I wouldn’t actually call the examples I’ve seen of ControlNet to be examples of fine control. I’m struggling to find a way to clearly communicate it, but it’s like the difference between 3D art that is trying to look like 2D, and actual 2D.
There’s always something lost in the translation.
Most artistic disciplines are their own language, and I just don’t think we have a way to communicate that language without actually doing the art, and art requires artistic intent, which I don’t think is possible with the current AI tools. Maybe it will be at some point, but artistic intent and control over the process are so interconnected that the balance becomes very difficult.
AI art is antithetical to art. Art requires artistic intent.
It could have some limited application for very early exploration in commercial art, or perhaps as very limited tools used in existing art software, but generative art is inherently pointless and you need artists to be able to do incremental iterations properly, which is required for real work, which isn’t supported yet. I’ll sure it’ll get better and more convincing, but it’s still inherently pointless to use AI for art, since the is supposed to be human expression.
Until we have autonomous agents, AI art is human expression
I strongly disagree. 99% of the work is being done by an algorithm. It’s like if we had autonomous driving and you said you were actively driving all day, because you told the car where to go, and then took a nap in the car until you had arrived.
Photography is widely recognized as an art form, even though the scene exists independently and the photographer “simply” frames and captures the shot.
A better driving analogy might be Tesla’s current level of self-driving, where you have to keep a hand on the wheel and eyes on the road the whole time, and remain in charge of all the critical decisions. When someone arrives in a Tesla and says, “I drove here,” no one goes “ackchyually…” Even if we follow your analogy, it’s the individual’s idea to reach that destination- often a novel place no one has even been to before.
Creative individuals curate unique datasets, which can take countless hours of manual work, to create LoRAs. They often draw from their own photographs, drawings, paintings, etc., and then coordinate prompts and parameters to blend their custom LoRAs with other creators’ LoRAs/models/checkpoints to craft something unique. These creations exist only because they had the vision and put in the effort to realize it. The process can be even more involved with tools like ControlNet, where artists might even sketch an outline of the scene by hand.
A quick selfie might not be considered art, but intentional expression through creatively capturing a scene is (photography). Similarly, a quick generation via Copilot for a meme might not qualify as art, but intentional expression through creative generation certainly does.
I disagree with your analogy, as I find it overstates the active involvement of the driver (prompter) during the drive (actual image generation).
Preparation is it’s own process, whether you’re curating art you made yourself/stole from non-consensual artists, or have been finding references as an artist. Different skillset. They help the process of making the final image, but they are not a direct part of that process.
And let’s not kid ourselves about theses datasets. There’s no accountability so there’s no way to ensure that any dataset you’re getting from other people aren’t comprised of, at least partially, stolen art.
ControlNet let’s you add visuals to your prompt for greater control, but you’re still generating the image externally, and leaving the vast majority of the decision making to the model you’re using. Even if someone is happy with the result they get from a generative model and find it visually pleasant, that doesn’t make it art. The model is doing the work and the model cannot have artistic intent, so it cannot make art. It can make images and people can enjoy those, but those images aren’t something new.
They are amalgamations of most basic common denominator of existing things. It is much more like a really advanced collage that is great at hiding the seams.
Amazingly, everything you’ve written is wrong, but I really don’t care enough to further correct you
Nope.
When humans make art, they are constantly making decisions. Decisions, decisions, decisions. With every stroke of the pen, with every color (not just a generic pink, blue or yellow, but specific tones and shades of those), with every everything they to while making that piece, they are making a lot of micro-decisions. Those decisions are made in respect to the person that is making the art, as their personal life experiences are what dictate how they make such decisions, even if they don’t notice it.
AI art is not like that. With AI, you type a prompt and outcomes an image. The user does not have a say in any of the micro-decisions that when into making that piece. The AI it self isn’t making any decisions either, it is just making the mathematical weighted average of what images with a description with similar tokens look like, and simply copying said decisions. The AI does not decide, it simply regurgitates previous decisions.
You are looking at the lowest form of AI art and then declaring that to be all there is. There is more than just pressing generate to a prompt. If you’re actually interested, I wrote a length post about it: https://lemmy.world/comment/12743504
Yeah you definitely don’t know what you’re talking about
AI art is human expression in the same way that the Gaussian blur tool is. It’s a bunch of math spitting out a pattern based on specific inputs.
All while currently being as ethical as the fast fashion industry producing scam versions of high fashion products.
It has the potential to be very useful in certain applications, but right now, all it really does is create Content to be consumed. Kinda like elevator music or that horrible Corporate Memphis style that has invaded every piece of corporate media/advertising in recent years. Soulless and without meaning. It’s pretty high quality slop, all things considered, but slop nonetheless.
I think it’s safe to say you don’t know what you’re talking about
I love Ai defenders who are ready to tell you what art is and what artists wants. Like maybe instead of recomending this cloud based bullshit app, first try to pick up a pencil actually?
I understand the excitement, but it is very much a situation of a layman trying to describe to experts what the expert and all their peers need.
I think it is just because AI has been hyped so much, and has genuinely made such impressive progress that people get swept up by the excitement, and idea that they could make their ideas into something tangible. They just don’t know the amount of consideration that goes into translating that.
Right now AI art is like Google translate poems.
The majority of people using AI will not insist you use it, aside from just trying to get others a realistic look on what the technology actually is. Just like a photographer won’t preach to a painter that they should pick up a camera. But that does not mean there isn’t benefit to derive from understanding how the other produces art. And if the painter thinks the camera is doing all the work and the photographer is a fraud, it’s probably good for them to get some exposure and realize it’s not just pressing a button. Like I explain here how that metaphor works for AI.
Most people I know that use AI are *shocker*, artists from before AI was a thing. They know how to draw with pencils, brushes, sponges, but also painting programs like Photoshop, sculpting programs, modeling programs, surface painting programs, shader production, algorithmic art. They are through and through artists. Them adding AI to their toolbox does not change that.
It’s a tool. Artists will learn how to use it to create art.
Right now it is not a tool. Right now it is an attempt at replacing artists.
It could be implemented in existing softwares in parts to make it a useful tool. Like a tool that could easily recolour parts of a fully rendered illustration, while still respecting the artistic intent with the form and lightning.
But right now it just spits out the blandest stuff, based on what it has identified as the most common denominators in art.
A lot companies are using the AIl to attempt to replace artists, that does not mean that some artists are not using it like a tool already.
I know quite a few artists that are already training their own artwork into custom data sets.
People that say that AI could be used as a tool to help artists clearly as never pickup a pencil to draw. The thing that makes an artists voice, that makes that art theirs are the decisions they make while making their art.
When you are drawing something, you are constantly making small micro-decisions with every stroke of your pencil, and those decision and how you make them is what makes art so beautiful, as no two artists make those decision the same way and each artist as a certain consistency in those decisions that evolves with them as a person. As such, art is so much more than a pretty picture, it is a reflection of the person who made it. Those decisions are also the fun part of making art.
AI art doesn’t let you make any decisions: you type the prompt and out comes an image. An image made of an weighted average of human made images with a similar description. You have no say in the micro-decision the machine makes, you have no say on where exactly the pencil strokes go. Therefore this machine is useless for artists. You might say “Just edit the image!”, but that doesn’t help either, as editing the image still doesn’t give you that micro-level of decision making. Also, editing a flat image with just one layer is just as useful as any other image form any search engine image search result. Unlike text, which can be easily edited to be exactly what you want.
I know their might be some wait to integrate machine learning into art, but right now the tools available don’t do that.
Hi, person here who’s drawn extensively with pencil as a kid but then slowed down. AI has reignited my passion for art, because unlike pencil or even digital drawing, you can iterate much more quickly, which allows me to do so while working a demanding job and having a life besides it. You are severely underestimating the amount of micro-decisions that go into AI art, and more importantly AI assisted art. If you’ll allow me, let me explain.
Lets break it down into levels of effort and creative input, I like to refer back to photography since it has some comparison:
spoiler
spoiler
spoiler
spoiler
Sadly, most people that talk confidently about how much they hate AI just know point #1 and maybe point #2. But I see point #3 and #4. And when I talk to artists that haven’t yet picked up on AI, but if they are aware (or made aware) of #3 and #4, suddenly their perspective also changes in regards to AI. But the hostility and the blind anger makes it quite hard to get through to people that not all art with AI is made equally. We should encourage people that want to use AI to reach the point of #3 and #4 so that their creative input is significant to the point it’s actually something produced by their human creativity. And this is also where an existing artist switching to using AI will most likely start off from.
Also, in terms of time. #1 might take seconds, #2 might take minutes to hours, but #3 and #4 can take days or weeks. Still not as long as drawing those same pieces by hand, but sometimes I wish it was as easy as people would make it out to be.
I’m glad you’ve found value in it. I’ve played around with a similar workflow you describe in step 3 and 4, but I just find that it always produces the blandest version. Sometimes you get a surprising iteration, but I think being used to seeing visual patterns makes it much more obvious just how predictable the image details get. It just ends up looking like a compilation of techniques.
I’m sure it won’t bring the same results for everyone, and that’s fine. I’m glad it’s that way and not just so easy it would consume all other ways of making things, as it would cease to be a tool at that point and just become the entire process. I really just try to make what I am already imagining since before I even put anything to paper, and so I know what I want the AI to help me with ahead of time. I get surprisingly close to the idea in my mind because of that. Much closer than I reasonably could get in other ways with the finite time I have.
That sounds like a good tool for you then. I do art as my day to day, and there are definitely aspects of my work that I would love to have a bit of AI injection to help speed the process up, but that is much more as a tool directly integrated into the softwares I already use, like a beefed up content aware transform tool that allows me to move parts of a finished image around just a little bit, and having the AI fill in the small gaps that creates.
I see so many small ways to make the art process less painful or introduce more non-destructive editing tools, if only the AI was built into the software and actively training on the art you were doing, as you did it, rather than having it take over whole parts of the process as it is currently used.
Yes, I want it to go that way too. But that does take time. Which is why I think it’s a shame some people are really hostile to it indiscriminately. I get that the big corporations are absolutely wielding the worst aspects of it, and I hate that too. But once you go to grassroots and small corporations, suddenly there is a whole different view of it. AI is a product of humanity’s collective efforts, and as such all of humanity should be it’s benefactor, not a small group of people that just happened to have the most money and influence at the time. Especially if they had nothing to do with actually creating the technology behind it. And by making the story about them rather than the people using it as it should, it will inevitably lead to people falling behind to corporations, and widening the gap even further. Making artists even less valued than they already are. Which is something I’ve always fought against even before AI.
I mean, the hostility is entirely understandable. The current form of generative art is meant to replace artists. It is part of what is currently devastating peoples livelihoods, although I think some companies and clients are already learning that it currently leads to lower overall quality, due to how much harder it is to implement changes based on feedback. It lowers the overall quality bar, although it does have the potential to raise the floor a little. The larger models that are causing this hype are quite literally trained on the work of unwilling artists.
It is the most disrespectful and clearly ethically wrong basis to build it on, and it really begs the question of whether the ends justify the means. Beyond that, art is just not an area where we need AI. It largely hurts artists, is super energy demanding so it actively hurts the environment for no real benefit.
The energy would be so much better used solving actual problems, so more people could spend time doing things they enjoy. If some people enjoy AI generation, then that’s fine but I think it shouldn’t replace a passionate, skill-based workforce.
People who say that AI can’t be used as a tool to help artists clearly have never tried using AI as a tool. Everything you’ve written here is untrue.
Artists can manually curate unique datasets to create LoRAs. They can draw from their own photographs, drawings, paintings, etc., and then coordinate prompts and parameters to blend their custom LoRAs with other creators’ LoRAs/models/checkpoints to craft something unique. The process can be even more involved with tools like ControlNet, where artists can sketch an outline of the scene by hand. I.e., you can have precise control over where the pencil strokes go.
The tools available right now do that
I’ve used it and continue to research it.
You’re wrong. You can’t have the same level of control as the person above describes. Even if you train the model on your own work, it will still be the one generating every “stroke” of the pencil. If trained for it, it will do it based on how you must often do so, but you can’t clearly control it. You can’t control granular details of the process of creating the image. It’s all broad strokes. I don’t know what your level of experience with art is, but so much of what makes art is tied up in the process of having to think through every little addition you make to an image. And by little addition I don’t mean “let’s add a person here” but “let’s do these 200 individual strokes that make up that person”. The involvement in the process is the point, and when an image is generated for you, you remove so much of the involvement and granular decision making, that the actual point is lost.
It’s like cooking with premade, pre-prepared ingredients. You can pick the dish and put it together with the stuff you buy, but you can’t control the whole process, because you’ve given up that for the sake of speed and convenience, and the dish will be different for it.
With ControlNet, you can control that very finely though. It partially combines computer graphics into the mix which is definitely not easy to get into.
But it’s different from drawing, that is for sure. Different decisions with different outcomes and different possibilities. But you see similar differences with every form of art. You don’t make 200 strokes for a photograph, or for a cut out collage, or algorithmic art, or sculpting. But you do make different decisions that are similar in nature. They shape the end product, and in a way no other person would do exactly the same. You still have to be involved in every step of the process, even if some steps are no longer done by yourself, they are replaced by other decisions.
Again, this is assuming they aren’t just clicking generate and calling it good to go, but if someone is using ControlNet they should be well above that, since the quality of blank generations is often not the best and demands refinement to anyone with an creative outlook.
This is why we shouldn’t confuse AI art, AI assisted art, and other forms of visual art, even if they all end up making an image as the end result. Something can be impressive when drawn, but mediocre if made with AI. Just like painting a scene isn’t the same as taking a photograph of that same scene, even if they end up being visually similar. Everything exists in context.
And yes, you are giving up some creative control for sake of convenience. But the question is how much. A painter that hand crafts their brushes will make a different painting had they used a pre-made brush. But we can agree I think that the creative control they lost by doing so is negligible. Artists generally lean on what they have produced before as references too. 3D sculptors can start with a cube too, but if they’re going to be making a person they will start with a mesh of a human figure, male or female depending on what they want. There is no shame in taking shortcuts in steps where it’s possible, even if it’s commendable if you don’t.
Art takes long and is expensive to produce, it would be unsustainable in this modern day to do everything from scratch every time. And as long as you focus on the parts where your decisions make the artistic intent happen, you can still make something unique and valuable in it’s own right.
That’s a cool visualisation of what kind of visual input you can feed into the process with ControlNet.
And it really makes it clear that what AI images is good for if communicating a general idea. I think comparing AI generated or Assisted images or videos to photography is probably the closest analogous medium we have, but I think AI images are stort if in-between that and more classical art. You have more control over the more technical aspects of the image, as you can alter those things with big strokes, but you’ve given up too much control to really infused it with artistic intent. Even when photography, where you are generally limited by reality, you can better infused artistic intent into the picture, because you carefully examine what makes that object of the picture unique. Even if you try to direct AI models, it limit their scope they will always add whether the most average expression of what they’re adding, because that what it looks for in the training; the commonalities/averages of whatever it was trained on.
Even ControlNet is just a way to claw back a little more control over the process. I wouldn’t actually call the examples I’ve seen of ControlNet to be examples of fine control. I’m struggling to find a way to clearly communicate it, but it’s like the difference between 3D art that is trying to look like 2D, and actual 2D. There’s always something lost in the translation.
Most artistic disciplines are their own language, and I just don’t think we have a way to communicate that language without actually doing the art, and art requires artistic intent, which I don’t think is possible with the current AI tools. Maybe it will be at some point, but artistic intent and control over the process are so interconnected that the balance becomes very difficult.
Artists actually know it’s bullshit, not a tool