> correctly manipulating images (via their source code) without having been programmed for anything of the kind, I've read plenty of suspicions that it is mult

> correctly manipulating images (via their source code) without having been programmed for anything of the kind, I've read plenty of suspicions that it is multimodal. I guess he's confirming it's not?