Most AI-generated images with photorealistic and 3D elements have obvious defects, but I’m curious if anyone’s done some analysis on the flat cartoon-style AI images. Cartoons, comics, and 2D artwork usually aren’t meant to be photorealistic, but I can tell something is off at a glance. What exactly is it?
Basically: signal-to-noise ratio
Cory Doctorow explains it better than I can: https://pluralistic.net/2025/03/25/communicative-intent/