LLMs should use "TK" or stable diffusion (and the like) so as not to get hung up on sequential words/thoughts and fill them in later instead of hallucinating filler.

I think this is a great idea.