I referenced all this yesterday, when I spoke of a new leap in complex artificial intelligence image generators, and therefore in the burst of images I excitedly make with them, but this leap might have finally scrambled my brain.
It is not a situation of "What can I do with this?" but rather "What can't I do with this?" Yes, there are things that this techno wonder can't do, things it gets wrong, things it does not do well, but they are hard to get to! And along the way, in the experiments mostly with past work, but also with new work, I am piling up so many amazing things to show you that I hardly know where to begin.
When I first began with this I worked with style changes of my photos, which were a leap better than earlier iterations, though not a total revolution, but I hardly understood the wealth of what I could be doing.
I'm not sure I understand it yet.
I started with image descriptions, characters, text, and famous people, and it was so good at all of that I had to try it out on my dioramas.
I'm currently stuck in the dioramas because they're pretty close to everything I've been aiming for and almost every one I try ends up in a delightful surprise. It's not perfect, but it is now possible the last bits could be done in edits, and even still, these are already far ahead of my earlier ones. And though I'm pretty sure I'm just scratching the surface here, I feel I'd better trot out some examples before I'm positively buried under my creations.



















