06-28-2023, 03:39 AM
The issue with AI is that it will by default give you a person standing there. I find that even if I put in some pose or action, it gets ignored a good deal of the time. That's what I see when I look up 'AI artists'. They're posting tons and tons of people standing around.
But if you have the skill to draw and paint then you can get the AI to do more interesting things with Img2Img or ControlNet. Non-artists are limited to what they can put in the text prompt or take from some other image. We can guide it a lot better.
Granted, with all the ethical & potential legal issues currently, I'd say to use it as reference. Someday though, there will be a Stable Diffusion that runs on legally acquired images and it'll still be super powerful and that's when the art community will adopt this stuff for sure. So learning it and using it a bit for now will put you ahead of the curve.
But if you have the skill to draw and paint then you can get the AI to do more interesting things with Img2Img or ControlNet. Non-artists are limited to what they can put in the text prompt or take from some other image. We can guide it a lot better.
Granted, with all the ethical & potential legal issues currently, I'd say to use it as reference. Someday though, there will be a Stable Diffusion that runs on legally acquired images and it'll still be super powerful and that's when the art community will adopt this stuff for sure. So learning it and using it a bit for now will put you ahead of the curve.