Ah yes, because LLMs are so reliable at understanding nuance and social context.
Human reddit moderators already can't do that from my experience
I don’t know why you are downvoted. Reddit did annonce to some moderators that AI moderation was coming this year and I heard that some already have access for testing. I also heard that it works pretty well to automate many actions, though it still requires some humans.