• leanleft@lemmy.ml
    link
    fedilink
    English
    arrow-up
    0
    ·
    edit-2
    10 months ago

    u could also try making small valid changes to detail over time until the posted content is unrecognizable and probably** false.
    eg: i drive a red car
    i drive a blue car
    i drive a blue truck
    i dont drive anything

    those are all possibly true. but alot of them are incorrect.

    • nucleative@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      ·
      10 months ago

      A python script hooked up to the openai API could be a fun way to play with this. Just edit the comments with random bs somehow marginally related to the original topic but incorrect.

      It’s a real shame though because those old comments are often lifesavers when you’re looking into really niche subjects.