in reply to Re: Incorporating ChatGPT into PerlMonks
in thread Incorporating ChatGPT into PerlMonks

This node falls below the community's minimum standard of quality and will not be displayed.
  • Comment on Re^2: Incorporating ChatGPT into PerlMonks

Replies are listed 'Best First'.
Re^3: Incorporating ChatGPT into PerlMonks
by dsheroh (Monsignor) on Oct 15, 2024 at 07:53 UTC
    That's a very... optimistic... view.

    It seems to me that a more likely "if it did say something stupid" scenario is that the rookie posting the question would instantly receive a bad answer, then run off to try to implement it without waiting for people who "want to post something quick to correct the wrong answer" to vet the answer for quality. Rookie wastes a lot of time and may actually cause some damage while attempting to implement the LLM's wrong answer. Perhaps they then return to PM and find the corrections (at which point they might be legitimately angry at PM having wasted their time with the immediate crap answer) or perhaps they don't, and they look elsewhere, and they decide that PM is a shit site giving shit answers that waste people's time.

    More people writing more correct solutions is indeed a good thing, but trying to prompt that by giving rookies bad answers first, so that experts will then show up to post corrections, is a piss-poor way of going about achieving that.