I recommend you try it before judging. I will be honest it has been actually very useful.

You are also assuming that the LLM is providing false information (or will have a higher chance of providing false information than a human)