Netflix has a great movie right now, Subservience. I was depressed after the first 15 mins where we see a bunch of construction workers and tradesmen fired and replaced by robots with no indication or idea of what they’re supposed to do afterwards, just replaced, bye, don’t care if you survive. They’re starting with programmers, white collar jobs like accounting, BI, tech support. Then they’ll move onto skilled trades. Subservience doesn’t sound so far fetched, so how long do we have?

  • garretble@lemmy.world
    link
    fedilink
    English
    arrow-up
    0
    ·
    9 days ago

    genAI is currently very stupid, and each new version is having diminishing returns on what the models can provide.

    I’m really not that worried, though some jobs may shift to instead of simply doing the job the worker will be required to fix all the mistakes the genAI came up with.

    • DuckWrangler9000@lemmy.worldOP
      link
      fedilink
      English
      arrow-up
      0
      ·
      9 days ago

      genAI is currently very stupid, and each new version is having diminishing returns on what the models can provide.

      Objectively false statement. You can dislike it (which I do too), but you’re totally wrong. AI passed the bar, AI can identify cancer images, AI can do things which very intelligent people can’t. Each new version has surpassed the previous, and companies have started layoffs to replace workers with skeleton crews armed with AI models.

      • FigMcLargeHuge@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        0
        ·
        9 days ago

        Software “passing” a written test where it has been fed all of the answers is not that impressive to me, except from a programming standpoint. From the intelligence standpoint, not so much. Just because it can pass the bar doesn’t make it lawyer worthy. This software cannot think, and that is ultimately what you pay a lawyer to do. None of this software is coming up with new thoughts, it’s just regurgitating what it has been fed.

  • Telorand@reddthat.com
    link
    fedilink
    English
    arrow-up
    0
    ·
    9 days ago

    A long time. AI ≠ AGI, and it’s prone to errors with even the most basic tasks. We already have robots that can do various things without AI, and they mostly replace workers doing tasks that are repetitive or dangerous, yet we still have plenty of factories and other jobs with human workers.

    On top of that, these robots need maintenance (including code), and they’re incredibly expensive. Replacing a human isn’t a simple drop-in, and they have a long way to go before they’re even close to parity with human cognition and context awareness.