I've found you still can't rely on LLMs to do anything 100% correct without human oversight. Unless you spend a lot of time prompt engineering and testing. Even then you might not get as close to 100% as you'd like.
But as you say, they are still extremely helpful anyway.
It's still extremely helpful, I just found it strange that it seemed like a simple task - for something that has been fed millions of documents, for it to still give some incorrect results - especially AFTER it had analysed its own results and found some noun artikels to be incorrect.