I got given an AI slop set of plans to sort out - admittedly with an apology and a “this doesn’t look right can you fix it” but eurghhh

A sample of the nonsense - I’m not sure I can make a 10nm brace, and I certainly don’t have any furby cactors

  • Canaconda@lemmy.ca
    link
    fedilink
    arrow-up
    12
    ·
    1 month ago

    I love how AI took the “add reddit to your search terms” literally and will now hallucinate facts based on a single reddit comment.

    • BCsven@lemmy.ca
      link
      fedilink
      arrow-up
      6
      ·
      1 month ago

      The best part of this is the comment sections where people are knowingly making up false stuff for /s or fun times, and the comments grow long with people adding to it. AI has no clue about the meta jokes and human quirky humour.

  • litchralee@sh.itjust.works
    link
    fedilink
    arrow-up
    8
    arrow-down
    1
    ·
    1 month ago

    Somehow, against all expectations, humanity has managed to exacerbate the XY problem in the context of woodworking, in novel and spectacular fashion, courtesy of AI generated slop.

    I have to imagine that if they gave you their original intention, or even the input prompt rather than the slopified output from an LLM, things would go a lot smoother. But alas, why should life get easier over time…

    • CelloMike@lemmy.worldOP
      link
      fedilink
      arrow-up
      7
      ·
      1 month ago

      Literally true

      The plan was full of odd choices so I had to have 5x the back and forth with the client just checking which things were required and which the machine had hallucinated, whereas even just a napkin sketch would’ve been easier since I’d know everything on there was intentional