Comment

Comment by ToucanLoucan

Not even deleted! Actively stolen. No one has yet grappled with the reality that especially in the realm of generative art, these models do not work without a basis of absolutely stressful amounts of stolen artwork to train them. They simply do. Not. Exist. Without that baseline unethical contribution that tons of artists made without their knowledge and certainly without their consent. None of the major figures behind any of this tech will acknowledge this. They make some hand-wavy non-committal gestures towards fair use and "how artists learn" and act like that's an answer, completely discarding tons of points not worth revisiting here because it's frankly been beaten to death already and if you're still fence sitting or pro generative art at this point, it's because you want/need to be to prevent cognitive dissonance in one way or another.

And to be clear: this stuff is SO COOL. The idea of entering prompts and getting at least vaguely unique and interesting output is incredibly cool. The notion of machines that learn, even in a restrictive way as the current models do, is so fascinating. And of course, like any cool and interesting new tech, it was immediately adopted/hijacked by the worst people imaginable, determined to fill Amazon self-publishing with generated hack kids books filled with generated text and generated art, at the lowest cost point possible, sold on the cheap, with the idea that they could make them convincing enough to trick some poor overworked parents into buying them and pocketing the cash. Just grift. Every new technology we get now, as few as the true new innovations they are, are always, always, immediately, co-opted by the worst actors in the space and turned to utter garbage.

Discussion

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton