4 ms·

Yes, you can debate the fairness angle. But an important angle is missed here. Apparently, this output, in whichever way it was constructed, was favored. Maybe

by fleddr 4y ago

Yes, you can debate the fairness angle. But an important angle is missed here. Apparently, this output, in whichever way it was constructed, was favored. Maybe it's just pretty, maybe it emotes, maybe it communicates something very well. I don't know, but I assume the jury was competent in being an art judge. My point is that this redirects a lot of talking points by artists to the trash bin. Art only has meaning when created by a human. NOPE. An artists' "lived experience" is the main driver of art. NOPE. Art is meaning and only humans can communicate meaning. NOPE. All these arguments are largely sentimental as actual output seems to be leading. I can already foresee elitist art critics pondering for days over the meaning of a piece, whilst sipping their wine. Whilst "the piece" is a guy in a basement typing "pretty fantasy landscape, epic lighting".

andsoitis 4y ago
> I assume the jury was competent in being an art judge. At a state fair? According to their website, this is the bio of the Exhibition Judge 2022 (they only show one judge but maybe there are more?): "Dagny McKinley is the Executive Director of the Colorado New Play Festival. Prior to that, she served for four years as Director of Development and Communications for Steamboat Creates. She is an author and photographer with five books to her credit. Her photography has been published across the United States. She earned an MFA in Creative Writing from Naropa University. Dagny currently serves on the Mineral Springs of Steamboat Springs Protection Committee with the Tread of Pioneers Museum, is on the Colorado Business Committee for the Arts advocacy committee, and is determined to see a performing art center in Steamboat Springs come to fruition. Dagny believes that anything can happen when you believe the impossible is possible." https://coloradostatefair.com/competitions/general-entry-fine-arts/fine-arts-exhibition/ https://coloradostatefair.com/competitions/general-entry-fin...
soneca 4y ago
I debate that a prize (decided by an art “judge”) represents anything of matter for the value of an art piece. Gatekeepers, curators and prizes are marketing tools, not necessarily relevant to the value of art. They are necessary for promoting art creation, that’s where their value is. They are not relevant, IMO, for attributing value to art.
- 0thgen 4y ago
  I could see that prize judges aren't _exactly_ correct on assessing meaning, but are they not even close? Like, their judgments are totally orthogonal to quality?
  - jakelazaroff 4y ago
    What is quality? Is this “better” than a Rothko or a Mondrian or a Duchamp? We’re talking about art; it’s inherently subjective.
    - 0thgen 4y ago
      I'm operating under the assumption that there is some construct of quality out there than can predict which art pieces will be deemed "better" than others. I just don't have a good description of what it is.
  - soneca 4y ago
    I think they are orthogonal to value. As for quality, you have to define quality by any specific dimension; that's what the judge is judging. They define what quality is for them and then judge based on it. If that quality generates value for a person depends on what weight that dimension that was used as quality metric matters or affects the viewer. If you have a clear perspective of what quality dimension a judge is analyzing and it matches your own attribution of value, then I say, in this case, the judgment is a good proxy of value for you. But this is in individual cases, not as a statement of the intrinsic value of said piece of art in general. I am using this writing to think, so it is not a formed thesis on art value and quality. But that's my perception of the world.
    - 0thgen 4y ago
      Fair enough, and I think your description that quality depends on an individual's own weighted dimensions of interest is spot on. But I'd also append what you said with: human individuals probably share a lot the same dimensions of interest (and probably close to the same weights). Plus, those dimensions/weights would probably become more refined as you gain experience evaluating+generating art. If there is such a thing as "quality", the key question would be whether professional art critics and experienced artists are any better at finding it.
0thgen 4y ago
Is it worth pointing out that the model (midjourney) is extrapolating from the textures, spatial arrangements, and image statistics embedded in human art that's already existing? Not saying humans are the only thing that can produce art, just that -- in this case at least -- human artists carved out the space the model could generate within (which is a massive accomplishment).
- fleddr 4y ago
  That's fully correct. The AI learned to draw based on work done by humans.
  - forgotpwd16 4y ago
    It can be said this is similar to how many artists take inspiration from others works.
polio 4y ago
I don't think these points have been proven as decisively as you feel. Whether it was verified or not, the provenance of a work of art has a significant effect on how a viewer receives it. So much of why the Mona Lisa matters is because of the cultural lore around it. People don't flock to it because of the wavelengths of light coming off of it, or its physical dimensions. I have a hard time imagining the same kind of mythology building around most AI work, at the moment, largely due to the low marginal cost of production. In the case of the state fair, the judges didn't know that AI had participated in the construction of the art. I would argue that a fair evaluation would require being honest about how it was made. It may have resonated with the judges because they were impressed that a human being used a digital tablet in Photoshop to produce these pixels, the implicit assumption they made when they evaluated it. Even if there is human ingenuity involved in creating visual hints and writing prompts, a judge will need to understand that process in order to properly appreciate that. We can advocate for philosophies of art that consider only the final product, but this doesn't mesh with how people empirically relate to art.
- fleddr 4y ago
  Most non-AI art also does not ever reach the level of mythology you speak of. There are scenarios where input may matter, such as in an art contest. Where the craft part is tested, not just output. In the vast majority of cases though where the public meets art (websites, movies, games, magazines), I don't think anybody cares about how it was made. It's output-driven. It's not art for art's sake, it's art as utility. Imagine your favorite writer. They have an incredible imagination and writing skills. But they can't draw or paint for shit. Now this writer uses their imagination to generate incredible illustrations to enrich the book. Illustrations that are relevant and beautiful. Does it matter that AI helped with it, if it got the job done? I don't think so.
obstroperous 4y ago
It doesn't change anything about dialogue around art. A human is still putting together "the piece." The process to achieve the end result is just commodified. Technology has been doing this since the beginning.

by fleddr 4y ago

andsoitis 4y ago

> I assume the jury was competent in being an art judge. At a state fair? According to their website, this is the bio of the Exhibition Judge 2022 (they only show one judge but maybe there are more?): "Dagny McKinley is the Executive Director of the Colorado New Play Festival. Prior to that, she served for four years as Director of Development and Communications for Steamboat Creates. She is an author and photographer with five books to her credit. Her photography has been published across the United States. She earned an MFA in Creative Writing from Naropa University. Dagny currently serves on the Mineral Springs of Steamboat Springs Protection Committee with the Tread of Pioneers Museum, is on the Colorado Business Committee for the Arts advocacy committee, and is determined to see a performing art center in Steamboat Springs come to fruition. Dagny believes that anything can happen when you believe the impossible is possible." https://coloradostatefair.com/competitions/general-entry-fine-arts/fine-arts-exhibition/ https://coloradostatefair.com/competitions/general-entry-fin...

soneca 4y ago

I debate that a prize (decided by an art “judge”) represents anything of matter for the value of an art piece. Gatekeepers, curators and prizes are marketing tools, not necessarily relevant to the value of art. They are necessary for promoting art creation, that’s where their value is. They are not relevant, IMO, for attributing value to art.

0thgen 4y ago
I could see that prize judges aren't _exactly_ correct on assessing meaning, but are they not even close? Like, their judgments are totally orthogonal to quality?
- jakelazaroff 4y ago
  What is quality? Is this “better” than a Rothko or a Mondrian or a Duchamp? We’re talking about art; it’s inherently subjective.
  - 0thgen 4y ago
    I'm operating under the assumption that there is some construct of quality out there than can predict which art pieces will be deemed "better" than others. I just don't have a good description of what it is.
- soneca 4y ago
  I think they are orthogonal to value. As for quality, you have to define quality by any specific dimension; that's what the judge is judging. They define what quality is for them and then judge based on it. If that quality generates value for a person depends on what weight that dimension that was used as quality metric matters or affects the viewer. If you have a clear perspective of what quality dimension a judge is analyzing and it matches your own attribution of value, then I say, in this case, the judgment is a good proxy of value for you. But this is in individual cases, not as a statement of the intrinsic value of said piece of art in general. I am using this writing to think, so it is not a formed thesis on art value and quality. But that's my perception of the world.
  - 0thgen 4y ago
    Fair enough, and I think your description that quality depends on an individual's own weighted dimensions of interest is spot on. But I'd also append what you said with: human individuals probably share a lot the same dimensions of interest (and probably close to the same weights). Plus, those dimensions/weights would probably become more refined as you gain experience evaluating+generating art. If there is such a thing as "quality", the key question would be whether professional art critics and experienced artists are any better at finding it.

0thgen 4y ago

Is it worth pointing out that the model (midjourney) is extrapolating from the textures, spatial arrangements, and image statistics embedded in human art that's already existing? Not saying humans are the only thing that can produce art, just that -- in this case at least -- human artists carved out the space the model could generate within (which is a massive accomplishment).

fleddr 4y ago
That's fully correct. The AI learned to draw based on work done by humans.
- forgotpwd16 4y ago
  It can be said this is similar to how many artists take inspiration from others works.

polio 4y ago

I don't think these points have been proven as decisively as you feel. Whether it was verified or not, the provenance of a work of art has a significant effect on how a viewer receives it. So much of why the Mona Lisa matters is because of the cultural lore around it. People don't flock to it because of the wavelengths of light coming off of it, or its physical dimensions. I have a hard time imagining the same kind of mythology building around most AI work, at the moment, largely due to the low marginal cost of production. In the case of the state fair, the judges didn't know that AI had participated in the construction of the art. I would argue that a fair evaluation would require being honest about how it was made. It may have resonated with the judges because they were impressed that a human being used a digital tablet in Photoshop to produce these pixels, the implicit assumption they made when they evaluated it. Even if there is human ingenuity involved in creating visual hints and writing prompts, a judge will need to understand that process in order to properly appreciate that. We can advocate for philosophies of art that consider only the final product, but this doesn't mesh with how people empirically relate to art.

fleddr 4y ago
Most non-AI art also does not ever reach the level of mythology you speak of. There are scenarios where input may matter, such as in an art contest. Where the craft part is tested, not just output. In the vast majority of cases though where the public meets art (websites, movies, games, magazines), I don't think anybody cares about how it was made. It's output-driven. It's not art for art's sake, it's art as utility. Imagine your favorite writer. They have an incredible imagination and writing skills. But they can't draw or paint for shit. Now this writer uses their imagination to generate incredible illustrations to enrich the book. Illustrations that are relevant and beautiful. Does it matter that AI helped with it, if it got the job done? I don't think so.

obstroperous 4y ago

It doesn't change anything about dialogue around art. A human is still putting together "the piece." The process to achieve the end result is just commodified. Technology has been doing this since the beginning.