• ToastedPlanet@lemmy.blahaj.zone
      link
      fedilink
      English
      arrow-up
      5
      ·
      3 days ago

      That’s basically the strategy, to make a long list of keywords. Any sentence structure is purely for the convenience of the prompter to be able to read and edit their own prompt. I found these AI generated videos the other day and they had a giant list of the keywords used in the video description.

    • edgemaster72@lemmy.world
      link
      fedilink
      English
      arrow-up
      6
      ·
      edit-2
      3 days ago

      I’m almost curious as to what it would come up with for that prompt, but I wouldn’t even know where to go to find out, and just figuring that out would take more energy than deserves to be expended on it, let alone the electricity needed

      • Rhaedas@fedia.io
        link
        fedilink
        arrow-up
        2
        ·
        3 days ago

        You’re right about the effort not being worth it, but as for experimenting all you need is a mid level or higher GPU and run ComfyUI, a flow-based image generator (and more, but it started as a Stable Diffusion alternative).

        As is, it’s a very broad prompt that will be far more likely to give terrible results. But it’s a comic, it isn’t meant to be that literal, and was obviously a jab at the idea of AI generation being art… which is its own debate and not as cut and dry as most think it is.