• flamingo_pinyata@sopuli.xyz
    link
    fedilink
    English
    arrow-up
    16
    arrow-down
    1
    ·
    29 days ago

    But how? The thing is utterly dumb. How do you even have a conversation without quitting in frustration from it’s obviously robotic answers?

    But then there’s people who have romantic and sexual relationships with inanimate objects, so I guess nothing new.

    • Opinionhaver@feddit.uk
      link
      fedilink
      English
      arrow-up
      10
      arrow-down
      1
      ·
      29 days ago

      How do you even have a conversation without quitting in frustration from it’s obviously robotic answers?

      Talking with actual people online isn’t much better. ChatGPT might sound robotic, but it’s extremely polite, actually reads what you say, and responds to it. It doesn’t jump to hasty, unfounded conclusions about you based on tiny bits of information you reveal. When you’re wrong, it just tells you what you’re wrong about - it doesn’t call you an idiot and tell you to go read more. Even in touchy discussions, it stays calm and measured, rather than getting overwhelmed with emotion, which becomes painfully obvious in how people respond. The experience of having difficult conversations online is often the exact opposite. A huge number of people on message boards are outright awful to those they disagree with.

      Here’s a good example of the kind of angry, hateful message you’ll never get from ChatGPT - and honestly, I’d take a robotic response over that any day.

      I think these people were already crazy if they’re willing to let a machine shovel garbage into their mouths blindly. Fucking mindless zombies eating up whatever is big and trendy.

      • musubibreakfast@lemm.ee
        link
        fedilink
        English
        arrow-up
        4
        ·
        29 days ago

        Hey buddy, I’ve had enough of you and your sensible opinions. Meet me in the parking lot of the Wallgreens on the corner of Coursey and Jones Creek in Baton Rouge on april 7th at 10 p.m. We’re going to fight to the death, no holds barred, shopping cart combos allowed, pistols only, no scope 360, tag team style, entourage allowed.

      • weedwolf@lemmy.world
        link
        fedilink
        English
        arrow-up
        2
        ·
        29 days ago

        I agree with what you say, and I for one have had my fair share of shit asses on forums and discussion boards. But this response also fuels my suspicion that my friend group has started using it in place of human interactions to form thoughts, opinions, and responses during our conversations. Almost like an emotional crutch to talk in conversation, but not exactly? It’s hard to pin point.

        I’ve recently been tone policed a lot more over things that in normal real life interactions would be light hearted or easy to ignore and move on - I’m not shouting obscenities or calling anyone names, it’s just harmless misunderstandings that come from tone deafness of text. I’m talking like putting a cute emoji and saying words like silly willy is becoming offensive to people I know personally. It wasn’t until I asked a rhetorical question to invoke a thoughtful conversation where I had to think about what was even happening - someone responded with an answer literally from ChatGPT and they provided a technical definition to something that was apart of my question. Your answer has finally started linking things for me; for better or for worse people are using it because you don’t receive offensive or flamed answers. My new suspicion is that some people are now taking those answers, and applying the expectation to people they know in real life, and when someone doesn’t respond in the same predictable manner of AI they become upset and further isolated from real life interactions or text conversations with real people.

    • Telorand@reddthat.com
      link
      fedilink
      English
      arrow-up
      8
      ·
      29 days ago

      In some ways, it’s like Wikipedia but with a gigantic database of the internet in general (stupidity included). Because it can string together confident-sounding sentences, people think it’s this magical machine that understands broad contexts and can provide facts and summaries of concepts that take humans lifetimes to study.

      It’s the conspiracy theorists’ and reactionaries’ dream: you too can be as smart and special as the educated experts, and all you have to do is ask a machine a few questions.

    • saltesc@lemmy.world
      link
      fedilink
      English
      arrow-up
      3
      ·
      29 days ago

      Yeah, the more I use it, the more I regret asking it for assistance. LLMs are the epitome of confidentiality incorrect.

      It’s good fun watching friends ask it stuff they’re already experienced in. Then the pin drops

    • glitchdx@lemmy.world
      link
      fedilink
      English
      arrow-up
      2
      ·
      29 days ago

      The fact that it’s not a person is a feature, not a bug.

      openai has recently made changes to the 4o model, my trusty goto for lore building and drunken rambling, and now I don’t like it. It now pretends to have emotions, and uses the slang of brainrot influencers. very “fellow kids” energy. It’s also become a sicophant, and has lost its ability to be critical of my inputs. I see these changes as highly manipulative, and it offends me that it might be working.

    • Victor@lemmy.world
      link
      fedilink
      English
      arrow-up
      1
      ·
      29 days ago

      At first glance I thought you wrote “inmate objects”, but I was not really relieved when I noticed what you actually wrote.