What about it is extreme? It's a concept, like "generate an xray image" eventually hopefully the cure to cancer could be represented as a simple molecule or whatever, I'm not saying I know.

There is currently no knowledge nor progress for what a cure for cancer, and nothing a LLM can draw upon.

You could generate "pregnant Elon Musk with four arms and three eyes doing yoga poses" because the image models have enough visual concepts of each of those individual things, but that specific image is (likely) not in any training dataset.