At a conference alongside Elon Musk, Nvidia CEO Jensen Huang suggested that any AI productivity gains will leave us working more, not less.
  • shalafi@lemmy.worldEnglish
    10·
    6 hours ago

    As usual, no one read the article. Huang didn’t say anything remotely like that. His idea is that AI will take work away, allowing us to pursue ideas we hadn’t had time for. Yes, we tend to work harder when pursuing our own ideas and plans. Nice theory, but we know it won’t work that way.

    monopoly capitalism

    That’s the word I’ve been needing! I keep saying what you are experiencing is not capitalism, it’s oligarchy. Monopoly capitalism is maybe more to the point.

    40-years ago was nothing like today, with a handful of companies running every sector of the economy and our lives. People in the 80s would shit kittens if you went back and explained who owns what in 2025.

  • frog_brawler@lemmy.world
    25·
    11 hours ago

    It’s already doing that. I have to refute AI bullshit from senior leadership on a daily basis now.

    Fuck him and fuck Nvidia.

  • Jared White ✌️ [HWC]@humansare.socialEnglish
    8·
    9 hours ago

    The sheer degree to which all of these moguls’ ideas are utterly bankrupt and devoid of real human craft & creativity continues to be shocking to me. Like, if any random person on the street got scooped up and put in front of a microphone and said the exact same things, everyone would think they were fucking idiots. Because they somehow are billionaires and are CEOs, they’re geniuses? Spare me, please.

      • cecilkorik@piefed.caEnglish
        9·
        8 hours ago

        Thank jebus there is somebody else in the world who sees this. I was literally playing with a modestly-large dataset Markov-chain based bot more than 25 years ago (yes in the late 90s) trained using data collected from a public chatroom (MOO) and it behaved exactly the same as these LLMs do today. A briefly convincing facsimile of a genuine conversation that quickly wandered off into utter nonsense sprinkled with occasional little nuggets of what seems like intelligence, wit or humor because we interpret it that way. It doesn’t know what the hell it’s saying. It never has.

        NONE of the fundamental flaws of the technology have been addressed. It does NOT have intelligence, it is simply a word predictor that, like a broken clock, is right twice a day, and they have just started adding layers and layers of filters on top while asking it over and over behind the scenes until they get a seemingly acceptable answer that passes all the filters. Sometimes that works well, sometimes it works really stupidly. It is random. That’s what randomness does. Given enough computations it will eventually produce the works of Shakespeare, but that does not make it a replacement for Shakespeare. It is not intelligent. It is random, and it is a toy.

        The only thing it is genuinely useful for is tricking humans into briefly thinking it is intelligent and correct and being pretty surprisingly convincing (which says more about us and our brains than it does about the tech), but the illusion should break eventually (and if it doesn’t you’re in big trouble, because it is still an illusion).

        The purveyors of this technology have tricked billions of people using only those features, and gaslit anyone who sees through it as “you’re just not prompting right” or “you’re not using our latest greatest tech that contains even more stolen human-made data it can copy-paste in to look like it knows what it’s talking about” and doing this has supposedly earned them trillions of dollars but it is just a trick. There’s nothing there. The money has been stolen and passed around in circles to inflated its value. There’s nothing there.

    • myrmidex@belgae.social
      5·
      11 hours ago

      Even more modern than good old surplus value extraction! We’ve been waiting for over 2 centuries now for these chains to be upgraded (again).