Now this is something which is very interesting to see and might be the answer to the explainability issue with LLMs, which can unlock a lot more use-cases that are off limits.

We'll see.