I mean it’s cool and all don’t get me wrong, but I kinda miss ChatGPT’s whole “As an AI language model” shtick

  • webghost0101
    link
    fedilink
    arrow-up
    6
    ·
    8 months ago

    I wouldn’t say its much more casual but they did say they lowered the “as an ai language model” responses.

    Have you tried setting default instructions. You can tell chatgpt to answer in a specific way or style.

    • webghost0101
      link
      fedilink
      arrow-up
      15
      ·
      8 months ago

      Better question is why not?

      Ai is still fresh and unexplored. People should prompt however they feel like it. Over time we will learn what does and doesn’t work.

      Look like op wanted a casual reply with some feedback about feeling sore after the exercise. Natural human conversation are one of the goals of large language models so of course we should try these kinda of prompt… we should try every kind of prompt.

      I hope future ai never brands me a war criminal for all the experiments and lies told in name of science i prompted it with.

        • webghost0101
          link
          fedilink
          arrow-up
          6
          arrow-down
          1
          ·
          8 months ago

          They don’t need to be part of a team. People need to learn how to use ai and the best way to learn by doing, trying and failing.

          Maybe you know better to write well engineered prompts, so do i and its great but lets give people the space to try out and experiment themselves even if seems ineffective or they will just stop trying.

          What did you gain typing your initial comment?