• Little8Lost@lemmy.world
    link
    fedilink
    English
    arrow-up
    20
    arrow-down
    3
    ·
    8 days ago

    I am definetly guilty even tho i should start judging after knowing if its useful ai or gen ai

    • audaxdreik@pawb.social
      link
      fedilink
      arrow-up
      28
      arrow-down
      8
      ·
      8 days ago

      I really don’t think there is any useful generative or general AI.

      So a lot of the issue is how marketing got their slimy tentacles around the word, but most “useful” AI is domain specific, symbolic ML (machine learning). Even LLMs have their uses in very specific domains, but again, general usage is very questionable.

      People are already somewhat familiar with ML, but that’s been kind of covered by the catch all term “algorithm”. What most people understand as “the” algorithm (YouTube, Twitter, whatever) isn’t a single algorithm, but a complex set of algorithms often at least partly compromised of some sort of ML.

      All that to say, the general public really doesn’t need to know this stuff and the serious engineers couldn’t care less of our opinion of it. Fuck AI.

      • Little8Lost@lemmy.world
        link
        fedilink
        English
        arrow-up
        3
        ·
        8 days ago

        There is ai used for enemys and video games and some for protein folding
        Thats the ones i mean with useful

        • audaxdreik@pawb.social
          link
          fedilink
          arrow-up
          3
          ·
          8 days ago

          Yeah, sorry, that wasn’t directed so much at you as it was using your post as a starting point.

          I remember the Folding at Home program, that was more about distributed computing than AI. Game AI has been well-discussed for decades now, but in 99.9% of other AI cases it’s usually in reference to the current trend (or trying to ride that wave) and like 0.1% niche nerd talk you caught a stray from.