It's art without an artist. Or maybe rather more accurately it's huge amounts of other peoples art put through a blender and reconstituted into a solid block of art product. Actually, yeah, it's the equivalent of spam, the tinned meat product. It has its uses, but it doesn't taste good to most people, and there are a few die hard fans.
Doing image editing/compositing/art yourself is just more genuine. It doesn't have to be the adobe product, there's tons of editors and art programs out there. Stumbling through gimp patching and editing a meme together gives the end result some charm imo. YMMV, but that's how I see it and I think a lot of others do to. Where AI generated imagery shines is utilitarian areas. Generating tokens for npcs in ttrpgs, rapidly iterating through concepts for a design, that sort of thing.
I draw and use AI in the same way I'll Google an image to get what my nonvisual memory is asking to see. Once I see it, I'm good to draw it.
FWIW, DALL-E2 was much much better than DALL-E3 at this task. It was enough to satisfy your imagination without being good enough to really look like "art". It was mass produced "art".
Yeah, but the machine learning is the art, not what it produces. Don't get me wrong, generative ai is super cool. However, justifiably or no, most people don't see someone who just types prompts into generative ai as an artist.