A thing I want to ask LLMs to do (in a series of prompts):

1) enumerate 100's of academic disciplines

2) for each list the 20 big ideas (replicated, core, differentiated ideas).

3) For each combination of insights from across domains, ask the LLM, "draw parallels and discuss any interesting overlap, and suggest further areas of insight this combined idea might imply".

Most combinations will be meaningless or uninteresting, but some will be game changing.

So far we've used humans swarming on large scale problems to find and try the interesting combinations.

But humans are expensive and also take time to come up to speed in a different discipline's jargon and knowledge, so we've tried only a tiny subset of combinations.

But LLMs have read all of the disciplines and also are cheap.

So as society we can now structurally find significantly more interesting and game-changing combinations.