I hate AI art and think it’s just soulless, derivative pixels
I look at it like the modern equivalent of newspaper political cartoons, or maybe as an evolution to the photoshopped placards on Late Night comedy news monologues.
Well, you can just feed AI a prompt and take the image that comes out, but that’s not how people do things anymore. AI art generation is now a complex set of image generation, in-/outpainting, tweaking, etc. I spent a couple of hours last night updating myself on how it is done, and I was shocked at all the changes that have taken place in the last six months. Now people are even passing their art through AI model subsets that they have trained themselves in order to get specific results, like specific backgrounds, vehicles, buildings… it’s incredible.
I don’t think we’re going to agree on this, but that’s okay :)
On one hand, I will agree, properly utilized, that it’s more like to a tool to be wielded than a “do it for me bot”; part of a larger process with the artist behind the wheel. On the other hand, though, I consider it the autotune of the visual medium (I’m not a fan of autotune in music, so at least I’m consistent?).
I’m definitely not saying “AI art isn’t art”, but I am saying that I am not a personal fan of it. It just ruins the magic for me.
I look at it like the modern equivalent of newspaper political cartoons, or maybe as an evolution to the photoshopped placards on Late Night comedy news monologues.
I mean, yeah, but a person who feels things, struggles, and understands the world still creates those cartoons/placards. Not so with generative AI.
Well, you can just feed AI a prompt and take the image that comes out, but that’s not how people do things anymore. AI art generation is now a complex set of image generation, in-/outpainting, tweaking, etc. I spent a couple of hours last night updating myself on how it is done, and I was shocked at all the changes that have taken place in the last six months. Now people are even passing their art through AI model subsets that they have trained themselves in order to get specific results, like specific backgrounds, vehicles, buildings… it’s incredible.
I don’t think we’re going to agree on this, but that’s okay :)
On one hand, I will agree, properly utilized, that it’s more like to a tool to be wielded than a “do it for me bot”; part of a larger process with the artist behind the wheel. On the other hand, though, I consider it the autotune of the visual medium (I’m not a fan of autotune in music, so at least I’m consistent?).
I’m definitely not saying “AI art isn’t art”, but I am saying that I am not a personal fan of it. It just ruins the magic for me.
Actually, “autotune of art” is about the best description of AI Art I’ve heard in a long time.
Call me old school, but I want to see where the artist made a happy little accident.