Found in NeMo.

  • AbouBenAdhem@lemmy.world
    link
    fedilink
    English
    arrow-up
    9
    arrow-down
    3
    ·
    edit-2
    9 months ago

    Seems to me like a reasonable criterion would be to determine if the trained model outputs copyright-infringing text in response to non-infringing prompts.

    • ryathal@sh.itjust.works
      link
      fedilink
      English
      arrow-up
      4
      ·
      9 months ago

      The ability to create a copy of a copyrighted work from ai shouldn’t be any more relevant than with a copy machine.