How early in a compounding process do you find the error?

If it's late, it can be super-linearly disruptive.

We haven't had too many computing-driven compounding processes before.

If it's a set of humans doing it, one of them will notice at some point and go "wait, what?"

But when an LLM is "yes, anding" its own output ^[cd]with no curation from someone with external judgment, it can get into an auto-catalysing rat's nest of confusion.