The thing is, LLMs are not-even-wrong. They are detached from correctness because they're answering a question that is loosely correlated with correctness. It'

The thing is, LLMs are not-even-wrong. They are detached from correctness because they're answering a question that is loosely correlated with correctness. It's not like they're attempting to be correct, or care about correctness