• Diurnambule@jlai.lu
    link
    fedilink
    English
    arrow-up
    4
    ·
    6 hours ago

    I wonder what would happen if some start to watermark document they doesn’t want in Claude training

    • turtlesareneat@piefed.ca
      link
      fedilink
      English
      arrow-up
      1
      ·
      2 hours ago

      There’s a key involved that we don’t have, so people can’t do this on their own. It’s pretty fascinating. Training models would have to be told to check for watermarks and ignore them, but yeah that would be an effective way for the providers to avoid ingesting their own AI output.