If you asked all the LLM to find flaws in the arguments presented, and to come up with counter-arguments, I doubt many current models would be so bad as to come up with that, and even the ones that did would fold when asked "how is that even an argument?".
alt Hacker News
If you asked all the LLM to find flaws in the arguments presented, and to come up with counter-arguments, I doubt many current models would be so bad as to come up with that, and even the ones that did would fold when asked "how is that even an argument?".