OpenAI now tries to hide that ChatGPT was trained on copyrighted books, including J.K. Rowling’s Harry Potter series::A new research paper laid out ways in which AI developers should try and avoid showing LLMs have been trained on copyrighted material.

  • Gnubyte@lemdit.com
    link
    fedilink
    English
    arrow-up
    16
    arrow-down
    19
    ·
    1 year ago

    Our ancient legal system trying to lend itself to “protecting authors” is fucking absurd. AI is the future. Are we really going to let everyone take a shot suing these guys over this crap? Its a useful program and infrastructure for everyone.

    Holding technology back for antiquated copyright law is downright absurd.

    • scarabic@lemmy.world
      link
      fedilink
      English
      arrow-up
      13
      arrow-down
      2
      ·
      1 year ago

      Stop comparing AI to a person. It’s not a person, it doesn’t do the things a person does, and it doesn’t have the rights of a person.

      And yes the laws are antiquated. We need new laws that will protect authors.

      Finally, no, you can’t just throw out all other considerations because you think AI is useful.

      • kava@lemmy.world
        link
        fedilink
        English
        arrow-up
        3
        arrow-down
        2
        ·
        1 year ago

        and it doesn’t have the rights of a person.

        And we have determined that AI created work cannot be copyrighted - because it’s not a person. Nobody’s trying to claim that AI somehow has the rights of a person.

        But reading a bunch of books and then creating new material using the knowledge gained in those books is not copyright infringement and should be not treated as such. I can take Andy Warhol’s style and create as many advertisements as I want with it. He doesn’t own the style, nobody does.

        Why should that be any different for a company using AI? Makes no sense to me.

        You have been duped into thinking copyright is protecting authors when really copyright primarily exists to protect companies like Disney.

      • Gnubyte@lemdit.com
        link
        fedilink
        English
        arrow-up
        1
        ·
        1 year ago

        I’m not sure about that at all. At what point does a computer program become intelligent enough to not have human rights but have some cognition of fair use.

        I think it needs to be really hashed out by someone who understands both copyright law and data warehouses, and some programming. It’s a sparse field for sure but we need someone equipped for it.

        Because I don’t think it’s as linear as you’re describing it.

    • CluckN@lemmy.world
      link
      fedilink
      English
      arrow-up
      5
      arrow-down
      2
      ·
      1 year ago

      People see that they were purchased for 10 billion dollars and want a piece of the pie.

    • LordShrek@lemmy.world
      link
      fedilink
      English
      arrow-up
      1
      arrow-down
      1
      ·
      1 year ago

      if I read a book no one sues my brain for consumption

      yes, this is the fundamental point

    • stappern@lemmy.one
      link
      fedilink
      English
      arrow-up
      2
      arrow-down
      6
      ·
      edit-2
      1 year ago

      also fuck authors. once you pass a million dollar profit you should lose any copyright claim