• scintilla@crust.piefed.social
    link
    fedilink
    English
    arrow-up
    5
    arrow-down
    1
    ·
    22 hours ago

    Hot take I know. I’m not talking about LLMs obviously but AI could absolutely be implemented to reduce risk in a nuclear plant. Having a constant double check running for every decision could reduce the risk of a tired human pressing the wrong button and leading to a SCRAM at the very least.

    • velindora@lemmy.cafe
      link
      fedilink
      English
      arrow-up
      9
      ·
      22 hours ago

      AI is great when it just watches and says “this is weird, maybe look at it” you can never have too many eyes

      • Lka1988@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        3
        ·
        18 hours ago

        That is an appropriate usage of “AI”, as it’s basically just pattern recognition. Something computers are really good at.

        • velindora@lemmy.cafe
          link
          fedilink
          English
          arrow-up
          1
          arrow-down
          1
          ·
          17 hours ago

          Not according to most of the people on Lemmy. They would have a nuclear meltdown (literally and figuratively) before allowing a computer program labeled “AI” to identify risk.