Early AI images had an eerie feeling about them. You could tell what the depicted objects were, yet the details were still unclear and foggy, like in a dream. It gave way for imagination and interpretation, and worked great for certain media. I think Warakami made a creepy visual story about visiting abandoned radioactive mall (where some deadly cult used to reside). He (she?) used early AI-generated pictures as illustrations, and it was perfect for creating uneasy atmosphere.
As for modern AI pics, they do feel different. I can't quite put my finger on it, but I think it has something to do with ratio of clear details vs unclear details. For the old ones, the ratio was about 50/50, while modern AI images are mostly realistic, yet with tiny but consistent mistakes that irritate the eye.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
As for modern AI pics, they do feel different. I can't quite put my finger on it, but I think it has something to do with ratio of clear details vs unclear details. For the old ones, the ratio was about 50/50, while modern AI images are mostly realistic, yet with tiny but consistent mistakes that irritate the eye.