Just as importantly: we have to stop amplifying tech bosses’ chosen narratives about their products’ destructive potential, because this helps them raise more money and do more terrible things. Bernie Sanders needs to stop insisting that the US government should own 50% of the money-losingest corporations the world has ever seen and start talking about how they will not get a government bailout when their investment bubble bursts.

  • MalReynolds@slrpnk.net
    link
    fedilink
    English
    arrow-up
    7
    ·
    edit-2
    22 hours ago

    It’s very capable of doing tasks where you’re fine with some failures and sub optimal results

    If you spend heaps of tokens (hey, new crypto) taking multiple runs on topics that can be easily verified as correct, preferably automatically, like security flaws and mathematical theorems. For everything else it’s technical debt, social decohesion, totalitarian auto-enforcement without mercy, billionaires wet dreams of worlds without workers and any number of other ‘hallucination’ driven misadventures.

    Pretty sure, if we last long enough, they’ll end up requiring a license to use, they seem as destructive, and yet maybe nearby as useful as cars were in specific cases (and IMO cars turned out pretty bad). As they stand now, only useful in the hands of someone already competent in a given field. Everyone else is in the ‘a little knowledge is a dangerous thing’ zone playing with TNT.

    • im_fine_sandy@nord.pub
      link
      fedilink
      English
      arrow-up
      1
      ·
      10 hours ago

      only useful in the hands of someone already competent in a given field

      Agreed. The problem being, how does someone become competent if they rely on LLMs to do things.