• @papertowels
    link
    1
    edit-2
    4 months ago

    You still need to be able to artistically judge that medium.

    That makes sense! I’d guess that the other guy agrees with this sentiment - their definition seems to center around art enabling the production process.

    It’s been a while since I’ve used it, but when I first started playing around with stable diffusion, the results were far from perfect. I’d have to regenerate segments multiple times, go back in with manual image editing, etc. before I would have something I was happy with. Generative models definitely aren’t “push button, receive art”. On the other hand, I’ve gotten a few pieces I rather enjoy, but I’ve had very little artistic training, so I’m grateful.

    • @barsoap@lemm.ee
      link
      fedilink
      24 months ago

      I’d have to […] go back in with […] editing, etc. before I would have something I was happy with.

      There. Artistry. The tension between vision and perception, the space of choices that has to be navigated.

      Most people’s vision when using SD is limited to some thematic idea and a pedestrian sense of aesthetics once they get something that even remotely hits both aspects they post it and call it art that’s a) not actually making choices and b) the reason why so much AI-generated stuff out there is barrel bottom at best.