In a way, sure. What’s unfortunate with such a medium is what a small proportion of the thoughts in the final product are the prompter’s. The machine references countless works that the prompter has no knowledge of, whereas in a medium controlled by the artist, those references (both conscious and subconscious) add meaning to the piece.
What’s unfortunate with such a medium is what a small proportion of the thoughts in the final product are the prompter’s. The machine references countless works that the prompter has no knowledge of, whereas in a medium controlled by the artist, those references (both conscious and subconscious) add meaning to the piece.
Here’s the part where I get mad at people in the thread that say I have no understanding of art (history). There’s been several ways of creating art that have been discussed that have the artist leave aspects to chance. It’s been done by Cage, people using radios in musical performances, introducing animals into artworks, using the brush like an idiot (Pollock) to achieve these things; in more modern mediums, the entire genre of rougelikes rely on chance to have certain things not be completely fixed, and emergent behaviour is a valuable aspect of creating interactive worlds that contain automatons.
Is there too much uncurated content in a Gen-AI work? Depends how much effort the creator put into the work, and what they are trying to do with it. I’m pretty sure for the actual scenario at the heart of the discussion, shitposts and strawberry textured elephants can have a few “happy accidents” without harming the central message… of the strawberry elephant.
The difference here is that an artist has control over the medium. Every letter was put there with intent, every stroke carries meaning. Deciding not to do these things can also carry weight, and even the decision to let chaos decide is a choice.
GenAI isn’t that, it removes the creative process entirely. Sure, you can get creative with prompt engineering, but the resulting art is the prompt not the AI generation.
It doesn’t matter how much work you put into micromanaging an artist, a commission is not your art. Similarly, it doesn’t matter how intricate and elegant your prompt is, you did not generate the result.
The difference here is that an artist has control over the medium. Every letter was put there with intent, every stroke carries meaning. Deciding not to do these things can also carry weight, and even the decision to let chaos decide is a choice.
no they don’t. That’s my whole point that there’s artworks where the artist doesn’t have complete control over the end result, and it’s the point.
It’s been done by Cage, people using radios in musical performances, introducing animals into artworks, using the brush like an idiot (Pollock) to achieve these things; in more modern mediums, the entire genre of rougelikes rely on chance to have certain things not be completely fixed, and emergent behaviour is a valuable aspect of creating interactive worlds that contain automatons.
And I would gladly take pollock out of the artistic cannon if I could.
In a way, sure. What’s unfortunate with such a medium is what a small proportion of the thoughts in the final product are the prompter’s. The machine references countless works that the prompter has no knowledge of, whereas in a medium controlled by the artist, those references (both conscious and subconscious) add meaning to the piece.
🤝
Here’s the part where I get mad at people in the thread that say I have no understanding of art (history). There’s been several ways of creating art that have been discussed that have the artist leave aspects to chance. It’s been done by Cage, people using radios in musical performances, introducing animals into artworks, using the brush like an idiot (Pollock) to achieve these things; in more modern mediums, the entire genre of rougelikes rely on chance to have certain things not be completely fixed, and emergent behaviour is a valuable aspect of creating interactive worlds that contain automatons.
Is there too much uncurated content in a Gen-AI work? Depends how much effort the creator put into the work, and what they are trying to do with it. I’m pretty sure for the actual scenario at the heart of the discussion, shitposts and strawberry textured elephants can have a few “happy accidents” without harming the central message… of the strawberry elephant.
The difference here is that an artist has control over the medium. Every letter was put there with intent, every stroke carries meaning. Deciding not to do these things can also carry weight, and even the decision to let chaos decide is a choice.
GenAI isn’t that, it removes the creative process entirely. Sure, you can get creative with prompt engineering, but the resulting art is the prompt not the AI generation.
It doesn’t matter how much work you put into micromanaging an artist, a commission is not your art. Similarly, it doesn’t matter how intricate and elegant your prompt is, you did not generate the result.
no they don’t. That’s my whole point that there’s artworks where the artist doesn’t have complete control over the end result, and it’s the point.
And I would gladly take pollock out of the artistic cannon if I could.