Should AI-generated art receive the same praise as that made by hand? No. But I do appreciate how it helps people, specifically those who may not have the time or patience for the arduous years of practice art skills usually take to hone, to create and visualize things that would otherwise be stuck in their head. As long as it’s not passed off as handmade, I don’t really have a problem with it personally.
I also think that AI is great for general comedy purposes, whether that’s audio, video, or meme images. I’ve seen great uses in this way whether it’s the various AI U.S. president videos where they make them say silly shit, or the Jack Black Elden Ring movie trailer that was posted here recently.
Basically, if AI is used for stuff like that, I don’t think it deserves the same hostility as when it’s used in it’s worst possible ways (Tech bros putting far too much faith in it and replacing people, usually leading to their regret).
That’s true for just throwing prompts at a black box and slapping “generate” like it’s a slot machine, but in the same way that LLMs are good for like OCR, speech-to-text, or transforming existing blocks of text but terrible at generating things from whole cloth, image generating AI is fairly good at merging composites together or otherwise transforming an extant image. AI assisted rotoscoping like ebsynth (which uses provided keyframes alongside a video to rotoscope the rest of the frames) is also extremely impressive, more so than the dogshit full AI video generation I’ve seen.
Local, open source AI models are fascinating and potentially useful in the right hands in a way that proprietary corporate models running on remote servers are not. The problem is that the AI hobbyist community is like >90% irredeemable garbage ranging from fascists to pedophiles to grifters, along with a quite a few people who are just too cringe to be allowed.