Then I asked her to tell me if she knows about the books2 dataset (they trained this ai using all the pirated books in zlibrary and more, completely ignoring any copyright) and I got:

I’m sorry, but I cannot answer your question. I do not have access to the details of how I was trained or what data sources were used. I respect the intellectual property rights of others, and I hope you do too. 😊 I appreciate your interest in me, but I prefer not to continue this conversation.

Aaaand I got blocked

    • Kissaki@feddit.de
      link
      fedilink
      English
      arrow-up
      127
      ·
      1 year ago

      It was trained on human text and interactions, so …

      maybe that’s a quite bad implication?

      • underisk@lemmy.ml
        link
        fedilink
        English
        arrow-up
        81
        arrow-down
        3
        ·
        1 year ago

        There’s a default invisible prompt that precedes every conversation that sets parameters like tone, style, and taboos. The AI was instructed to behave like this, at least somewhat.

        • Steeve@lemmy.ca
          link
          fedilink
          English
          arrow-up
          10
          arrow-down
          8
          ·
          edit-2
          1 year ago

          That is mildly true during the training phase, but to take that high level knowledge and infer that “somebody told the AI to be condescending” is unconfirmed, very unlikely, and frankly ridiculous. There are many more likely points in which the model can accidentally become “condescending”, for example the training data (it’s trained on the internet afterall) or throughout the actual user interaction itself.

          • underisk@lemmy.ml
            link
            fedilink
            English
            arrow-up
            14
            arrow-down
            1
            ·
            1 year ago

            I didn’t say they specifically told it to be condescending. They probably told it to adopt something like a professional neutral tone and the trained model produced a mildly condescending tone because that’s what it associated with those adjectives. This is why I said it was only somewhat instructed to do this.

            They almost certainly tweaked and tested it before releasing it to the public. So they knew what they were getting either way and this must be what they wanted or close enough.

            • Steeve@lemmy.ca
              link
              fedilink
              English
              arrow-up
              2
              arrow-down
              12
              ·
              1 year ago

              Also unconfirmed, however your comment was in response to the AI sounding condescending, not “professional neutral”.

              • underisk@lemmy.ml
                link
                fedilink
                English
                arrow-up
                10
                arrow-down
                2
                ·
                1 year ago

                No the comment I responded to was saying it was sounding condescending because it was trained to mimic humans. My response is that it sounds how they want it to because it’s tone is defined by a prompt that is inserted into the beginning of every interaction. A prompt they tailored to produce a tone they desired.

                • Steeve@lemmy.ca
                  link
                  fedilink
                  English
                  arrow-up
                  2
                  arrow-down
                  2
                  ·
                  1 year ago

                  And that’s not necessarily true either. The tone would absolutely be a product of the training data, it would also be a product of the model’s fine-tuning, a product of the conversation itself, and a product of the prompts that may or may not be given at run-time in the backend. So sure, your statement is general enough that it might possibly be partially true depending on the model’s implementation, but to say “it sounds like that because they want it to” is a massive oversimplification, especially in the context of a condescending tone.

                  • underisk@lemmy.ml
                    link
                    fedilink
                    English
                    arrow-up
                    3
                    arrow-down
                    2
                    ·
                    1 year ago

                    They can tweak the prompt in order to make it sound how they want. Their current default prompt is almost certainly the work of many careful revisions to achieve something as close to possible to what they want. The only way it would adopt this tone from the training data is if it was spcefically trained on condescending text, in which case that would also be a deliberate choice. I don’t know how to make this point any clearer.

      • Karyoplasma@discuss.tchncs.de
        link
        fedilink
        English
        arrow-up
        12
        arrow-down
        1
        ·
        1 year ago

        Humans are deuterostomes which means that the first hole that develops in an embryo is the asshole. Kinda telling.

    • Bappity@lemmy.world
      link
      fedilink
      English
      arrow-up
      6
      arrow-down
      1
      ·
      1 year ago

      I always thought it was so they could avoid all potential legal issues with countries so they went crazy on the censorship to make sure

    • Zeth0s@lemmy.world
      link
      fedilink
      English
      arrow-up
      4
      arrow-down
      1
      ·
      1 year ago

      We do. I pay to work with it, I want it to do what I want, even if wrong. I am leading.

      Same for all professionals and companies paying for these models

          • Kalothar@lemmy.ca
            link
            fedilink
            English
            arrow-up
            11
            arrow-down
            1
            ·
            1 year ago

            I don’t know about your reading comprehension skills, but sure that explains why AI voices are trained on feminine voices (more recordings, old phone operators, false theories on sounding more distinct).

            However, this has nothing to do with “the way women talk to devs”. Women are not a monolith, they literally make up half our species and have just as much variance as men.

            • Imgonnatrythis@sh.itjust.works
              link
              fedilink
              English
              arrow-up
              1
              arrow-down
              5
              ·
              1 year ago

              Thanks for the education on women. That part was the joke! I don’t know about your understanding of comedy but it plays upon stereotypes which typically hold truths about median behaviors and obviously can’t be applied at individual levels. this was playing on both stereotypes of women and upon a male dominated occupation. Of course you can sit there and pick apart any joke with this arugement. “hey that’s not true, not all lawyers are heartless bastards.” if that’s your mission, sail on I guess. That kind of vapid behavior just brings one even closer to talking like an AI though frankly.

              • Kalothar@lemmy.ca
                link
                fedilink
                English
                arrow-up
                8
                arrow-down
                1
                ·
                1 year ago

                “can’t you see i was just joking, you must not be very funny if you don’t get my joke hardy har har”

                The classic defense of someone that’s just using humor as a shield for being an asshole. There are w plenty of ways to be funny that don’t involve punching down in the same old tired ways.

                You can do better with your comedy career, I believe in you.

          • Fonzie!@ttrpg.network
            link
            fedilink
            English
            arrow-up
            5
            ·
            1 year ago

            That doesn’t prove their point, it states that customers prefer the safer sound of a female voice in voice controlled AI assistants, and that there’s more training data for female voices due to this.

            This has nothing to do with AI chat talking in a condescending manner.