I would argue that skill is not that much the issue.
If I had a button that put the image on my head directly onto a canvas I would be the best artist of the world. For me at least art is not about the skill but about the vision.
And the problem with AI is that it greatly limits the vision. It’s almost impossible to truly move the image or the concept you have in your head to the canvas using thar technique. Because the model will produce the images it has been train for, not the image in your head.
Other manual techniques do allow people to put the image in their head onto the canvas much more accurately and with greater intention. Independently of the complexity of skill.
I think photography is the greatest example. As it implies to tweak values and push a button, it’s not something that requires carefully hold a brush or have painting skills. But the camera allows you to produce the images as you want to produce them, with great intention.
That’s also why it’s true what the user abobe is saying that the person who has a pipeline and spent hours tweaking things with ai tools will get better results than the one who just wrote “draw me a pretty girl”. Because they would have probably spent a lot of time for the model to output something similar to what they had in mind. But still the models, even the most advanced ones, are too limiting in many areas.
And the problem with Al is that it greatly limits the vision. It’s almost impossible to truly move the image or the concept you have in your head to the canvas using thar technique. Because the model will produce the images it has been train for, not the image in your head.
This one statement invalidates everything you said because you clearly have no idea how these tools actually function.
And honestly, I think this is one of the biggest problems with people who argue against what they call “AI.” They’re so blindly opposed to these tools that they won’t even take the time to learn how they work, what they’re capable of, or how people actually use them. They form an opinion first, then refuse to do the research necessary to see whether that opinion even holds water.
You don’t have to like AI. You don’t have to use it. But if you’re going to argue against it, you should at least understand what the hell you’re arguing against.
Using Anima as base model, with controlUI, trying loras, IMG2IMG with flux.
It’s very clear that the generation is limited to a dataset.
Even more clear when you use SDXL that basically works by tags.
But even with models which process natural languages you see the limitations imposed by the training data.
If you want to make a generic image that’s been done 10000 times it’s very easy. But if you want to do something very specific that the thing have not been trained to do… you are not going to make it. Unless you literally paint want you want and train a lora with it, which kind of defeats the purpose.
I think that’s a pretty good interpretation; really the only thing I have to add is that I think of all the artistic skills, the most fundemental is the one that lets you translate from your idea to a real, tangible creation.
I would argue that skill is not that much the issue.
If I had a button that put the image on my head directly onto a canvas I would be the best artist of the world. For me at least art is not about the skill but about the vision.
And the problem with AI is that it greatly limits the vision. It’s almost impossible to truly move the image or the concept you have in your head to the canvas using thar technique. Because the model will produce the images it has been train for, not the image in your head.
Other manual techniques do allow people to put the image in their head onto the canvas much more accurately and with greater intention. Independently of the complexity of skill.
I think photography is the greatest example. As it implies to tweak values and push a button, it’s not something that requires carefully hold a brush or have painting skills. But the camera allows you to produce the images as you want to produce them, with great intention.
That’s also why it’s true what the user abobe is saying that the person who has a pipeline and spent hours tweaking things with ai tools will get better results than the one who just wrote “draw me a pretty girl”. Because they would have probably spent a lot of time for the model to output something similar to what they had in mind. But still the models, even the most advanced ones, are too limiting in many areas.
At least that’s my take.
This one statement invalidates everything you said because you clearly have no idea how these tools actually function.
And honestly, I think this is one of the biggest problems with people who argue against what they call “AI.” They’re so blindly opposed to these tools that they won’t even take the time to learn how they work, what they’re capable of, or how people actually use them. They form an opinion first, then refuse to do the research necessary to see whether that opinion even holds water.
You don’t have to like AI. You don’t have to use it. But if you’re going to argue against it, you should at least understand what the hell you’re arguing against.
I have tried.
I spent several days on a comfyui workflow.
Using Anima as base model, with controlUI, trying loras, IMG2IMG with flux.
It’s very clear that the generation is limited to a dataset.
Even more clear when you use SDXL that basically works by tags.
But even with models which process natural languages you see the limitations imposed by the training data.
If you want to make a generic image that’s been done 10000 times it’s very easy. But if you want to do something very specific that the thing have not been trained to do… you are not going to make it. Unless you literally paint want you want and train a lora with it, which kind of defeats the purpose.
I think that’s a pretty good interpretation; really the only thing I have to add is that I think of all the artistic skills, the most fundemental is the one that lets you translate from your idea to a real, tangible creation.