I call bullshit. Zero of the image AIs would generate these pictures based on these headlines, the headlines are FAR too simple to generate these particular images.
I might be wrong but I was thinking these images and headlines were prediction made by the AI based on real news headlines.
As far as I know they are real headlines and the whole thing is just that they were "illustrated" via image AIs based on those headlines.
But it's misleading because if you put simplistic small prompts like those headlines into these image AIs they won't generate nearly anything so specific as the very stylized, very detailed images shown in the video. So someone is bullshitting to a degree, in that it wasn't just the headlines fed to the image AIs to get those images. They also put in their own description of what they wanted the image to look like specifically in terms of style and content.
(post is archived)