• Note to Congress: Please try to keep up

    Home » Forums » Newsletter and Homepage topics » Note to Congress: Please try to keep up

    Author
    Topic
    #2516075

    LEGAL BRIEF By Max Stul Oppenheimer, Esq. That’s a big ask. In a previous column, I explained why law always lags technology. To summarize, case law i
    [See the full post at: Note to Congress: Please try to keep up]

    5 users thanked author for this post.
    Viewing 3 reply threads
    Author
    Replies
    • #2516088

      I don’t understand why problem 1 is a problem at all:
      “if AI machines cannot be authors or inventors, their work cannot be copyrighted or patented at all”

      Somebody owns the AI machine. They copyright or patent whatever their machine authors or invents. Am I missing something ?

      Windows 10 Home 22H2, Acer Aspire TC-1660 desktop + LibreOffice, non-techie

      1 user thanked author for this post.
      • #2516355

        I would think the laws which assign patent and copyright rights to employers could be extended to make this point explicit. In short, since as of now AI is created by humans, anything AI creates which is original enough to qualify, should be held to the same standards as works and inventions people create while in the service of their employers. The biggest change I would foresee is if it were one day deemed that AI had created itself without human intervention. At that point, the AI itself would need to be treated as a person under the law. This would require revising laws and regulations. We are a long way from that happening.

        -- rc primak

        2 users thanked author for this post.
        • #2519272

          A perfectly rational approach. Unfortunately, both the Copyright Office and the Patent Office take a different view – in their view (supported at this point by two federal cases) only a human can create and creating the machine that creates doesn’t qualify. Higher courts may side with your view, or Congress may resolve the issue, but at this point legal authority rejects the approach. Which is not to say that is a good result.

          1 user thanked author for this post.
    • #2517156

      Oh yes, creators do know and do very much care that their copyrighted work is being scraped and used to train AI without their permission, and they don’t like it one bit.

      The biggest kerfuffle in progress is among visual artists.  But just within the past few weeks, archiveofourown.org (AO3) has had fanfiction authors’ accounts slamming shut left and right, going registered-user-only to prevent, at least, future works from being used without their consent.  For existing works, we figure it’s most likely too late.

      How do we know it’s happening?  Well, I’m not sure whether anyone can specifically identify text lifted from their own individual works.  But terms have recently started popping up in AI-generated text that are extremely specific to certain fanfiction subgenres (and that may or may not be something the person using the AI wanted to see, which is another can of worms unto itself.)

      Among numerous issues with this, it seems that various apps using this AI technology are not being created by the same people who set up the AI, but by developers who are licensing or leasing or otherwise obtaining permission to use it.  So they have no idea where the AI’s dataset came from and no control over what information it does or doesn’t include.  Copyright owners who don’t want their works used in this way in a particular app have no recourse.

      If there’s a way to contact the creators of the AI itself (or even identify who they are) and request that something be excluded, so far I haven’t come across it.

      i7-10700k - ASROCK Z590 Pro4 - 1TB 970 EVO Plus M.2 - DDR4 3200 x 32GB - GeForce RTX 3060 Ti FTW - Windows 10 Pro

    • #2517387

      Oh yes, creators do know and do very much care that their copyrighted work is being scraped and used to train AI without their permission, and they don’t like it one bit.

      The ministry of of justice in Israel declared that there are no copyright infringements when training AI using copyrighted material.

      • #2518654

        I would express to you what I think of the ministry of justice in Israel’s opinion on the matter, but I suspect it would not make it past the moderators.

        i7-10700k - ASROCK Z590 Pro4 - 1TB 970 EVO Plus M.2 - DDR4 3200 x 32GB - GeForce RTX 3060 Ti FTW - Windows 10 Pro

    • #2517448

      This will only be a problem until we achieve post-scarcity, often posited by SF writers, which will occur once we are able to easily get off planet Earth and spread throughout the solar system.   At that point we will have access to unlimited resources, so there will be no need for ownership when everything is free.  Patents go out the window then.

    Viewing 3 reply threads
    Reply To: Note to Congress: Please try to keep up

    You can use BBCodes to format your content.
    Your account can't use all available BBCodes, they will be stripped before saving.

    Your information: