• jmcs@discuss.tchncs.de
    link
    fedilink
    English
    arrow-up
    13
    ·
    edit-2
    2 months ago

    Before we even get close to have this discussion, we would need to have an AI capable of experiencing things and developing an individual identity. And this goes completely opposite of the goals of corporations that develop AIs because they want something that can be mass deployed, centralised, and as predictable as possible - i.e. not individual agents capable of experience.

    If we ever have a truly sentient AI it’s not going to be designed by Google, OpenAI, or Deepmind.

    • Pennomi@lemmy.world
      link
      fedilink
      English
      arrow-up
      6
      arrow-down
      1
      ·
      2 months ago

      Yep, an AI can’t really experience anything if it never updates the weights during each interaction.

      Training is simply too slow for AI to be properly intelligent. When someone cracks that problem, I believe AGI is on the horizon.

        • Pennomi@lemmy.world
          link
          fedilink
          English
          arrow-up
          1
          ·
          edit-2
          2 months ago

          Artificial General Intelligence, or basically something that can properly adapt to whatever situation it’s put into. AGI isn’t necessarily smart, but it is very flexible and can learn from experience like a person can.

  • cm0002@lemmy.world
    link
    fedilink
    English
    arrow-up
    6
    arrow-down
    3
    ·
    2 months ago

    Gemini in it’s current form? No, but it is a fair question to ask for the future

        • justOnePersistentKbinPlease@fedia.io
          link
          fedilink
          arrow-up
          11
          ·
          2 months ago

          Yeah, but it’s like fusion. It’s always 20 years away for the last 60 years.

          Realistically, as a dev who watched AI develop from cheap parlor tricks to very expensive and ecosystem crunching fancy parlor tricks that mangers think will replace all of their expensive staff who actually know how to design and create:

          Modern “AI” is fundamentally incapable of actual thought. They are very advanced and impressive statistical engines, but the technology is incapable of thinking at a fundamental level.

  • minorkeys@lemmy.world
    link
    fedilink
    English
    arrow-up
    8
    ·
    2 months ago

    If we lose perspective that computer systems are machines, we’re fucked. Stop personifying computer systems just because they make you feel things. JFC.

    “Many of you feel bad for this lamp. That is because you crazy [sic]. It has no feelings…”

  • stoly@lemmy.world
    link
    fedilink
    English
    arrow-up
    1
    ·
    2 months ago

    This could potentially be a concept in 100 years but is a stupid question for now.

    • Possibly linux@lemmy.zip
      link
      fedilink
      English
      arrow-up
      4
      arrow-down
      1
      ·
      edit-2
      2 months ago

      I wonder if Gemma is actually a white man

      It is saddly common for LLMs to be racist and biased against people of color so maybe they are all secretly white racist males

  • Rhoeri@lemmy.world
    link
    fedilink
    English
    arrow-up
    5
    arrow-down
    2
    ·
    2 months ago

    So… the headline answered the question and people still read the article?

  • gravitas_deficiency@sh.itjust.works
    link
    fedilink
    English
    arrow-up
    19
    ·
    2 months ago

    Ok so: Measure of a Man is one of my all time favorite Star Trek episodes, but come the fuck on. We are so, so far away from that. Maybe worry more about humans, right now, and the world we live in, instead of some nebulous fucking future that we won’t even goddamn reach if we don’t pay attention to, you know, humans and the world we live in.

  • zbyte64@awful.systems
    link
    fedilink
    English
    arrow-up
    9
    ·
    2 months ago

    Can our AI fall in love with a human? Scientists laughed at me when I asked them but I found this weird billionaire to pay me to have sex with his robot.

  • nandeEbisu@lemmy.world
    link
    fedilink
    English
    arrow-up
    8
    ·
    2 months ago

    I have a preconceived conclusion about my anthropomorphized view of a statistical model with some heuristics around it. People who know what they’re talking about say I’m wrong, but I need an idea for an article to write that people will read.