Funny, but also, the reliability of art as an honest signal of elevated human work, skill, insight, or experience is like an arbitrage everyone piles into to re

Funny, but also, the reliability of art as an honest signal of elevated human work, skill, insight, or experience is like an arbitrage everyone piles into to reap the benefits, and then it no longer represents what it initially did. It's like how we mostly produce music instead of perform it now, where you don't have to spend years becoming a skilled musician to produce music - and so it's not as rare, exceptional, or valuable anymore. Everything loses value/meaning over time, and art that represents a value from a prior time will have lost most of it's meaning. This is kind of what po-mo theory was about, where people can't tell the difference between the real and represented, what a copies of a copy are, and the relativity of individual experience. AI art is a kind of perfection or demonstration of some of that criticism. What AI art does is use recombinations of existing inputs to produce new symbols that are internally consistent with themselves, and they are only meaningful to us because a) we have accepted the input symbols as premises, and b) we have trained it on our beliefs about consistency that reinforce our beliefs about those symbols. The product of these models not so much "real" as they are just iterations over a finite symbol space. They are artifacts of the logic of an idea. However, it may be more powerful than we expected, as it's also a perfect meta-criticism for evaluating critical theories, most of which originated (ironically) in literary and art criticism. This dynamic is why you can use ML to produce papers in "theory" that are indistinguishable from acadeimc writing, because these, too, are not honest signals of elevated human thinking or skill, but rather, just self-referential, interally consistent artifacts of ideology whose only bearing on human experience is what we assign to it. Like AI art, they are not an indicator for reality. We can use them to model reality, but confusing the model with the real is full of some pretty subtle tricks. Anyway, it yields a pretty funny tautology that shows the limits of both art and theory, where arguably, to coin a deepity, if you can simulate it - it's not real.