Comments
Loading Dream Comments...
You must be logged in to write a comment - Log In
This one felt like a true collaboration rather than just a commission. By that I mean that I don't think the AI could have produced this image without very careful guidance and my digital collage to work from as a base image. As I think I've mentioned before, the Text to Image generator has difficulty with incongruity- it often can't quite figure out how to combine ideas that have never been combined in the stock pictures it was trained on. Luckily, through careful phrasing and the meticulous referal to the base image, I managed to coax something very close to what I wanted from the AI: an image that combines art deco structures like hotels and spiral staircases with lovecraftian, cosmic backdrops and deep sea imagery. I couldn't have produced such a coherent image without the AI and the AI couldn't have produced such an incongruous or bizarre one without me. Like I said, it feels like a true collaboration between man and machine rather than just me commissioning the AI to create something for me.