If the LLM is smarter than the person, the person cannot look at the output and judge to see whether it is correct.
True, but that's a people problem.
alt Hacker News
True, but that's a people problem.