I mostly just think this is a bad response to a real problem.
Attitude problems aside[0], if you lead with "Responses from Large Language Models like ChatGPT, Claude, or Gemini are not facts", most people are probably going to respond with some form of "well it said this one fact and I know for sure that one was right" and move on from this unconvinced of anything.
I'm not sure what a better approach is though. Honestly, "Don’t copy-paste something that a chatbot said and send it to someone as if that’s authoritative" feels like a better starting point. Another comment in this thread about asking to share the prompt and demonstrating how it can be manipulated could help. Or talking about LLM bias. I dunno.
P.S. what's up with those form inputs/submits for the good/bad uses?
[0]: "lmgtfy" or even "rtfm" seemed to fade after the novelty wore off and everyone realized it was a jerk move to use terms like that. Or maybe it's a generational thing?
Attitude problems aside[0], if you lead with "Responses from Large Language Models like ChatGPT, Claude, or Gemini are not facts", most people are probably going to respond with some form of "well it said this one fact and I know for sure that one was right" and move on from this unconvinced of anything.
I'm not sure what a better approach is though. Honestly, "Don’t copy-paste something that a chatbot said and send it to someone as if that’s authoritative" feels like a better starting point. Another comment in this thread about asking to share the prompt and demonstrating how it can be manipulated could help. Or talking about LLM bias. I dunno.
P.S. what's up with those form inputs/submits for the good/bad uses?
[0]: "lmgtfy" or even "rtfm" seemed to fade after the novelty wore off and everyone realized it was a jerk move to use terms like that. Or maybe it's a generational thing?