Essay Question #1:

Your law firm represents a well-known screenwriter, Naomi Jones, whose drama series aired on a major network between 2018 and 2022. Jones recently discovered that substantial portions of her scripts were used without her knowledge or consent to train a generative AI model (“the NeuroVerse Model”) developed by a major tech company—NeuroVerse AI (“NeuroVerse”).

Internal documents leaked from NeuroVerse reveal that over 10,000 copyrighted scripts were used—without individual licenses—as part of the model’s training dataset. This included final scripts, as well as earlier drafts stored on cloud-based collaborative platforms. Among them were multiple confidential drafts of Jones’ scripts, which had been uploaded to Ubiquitous, a popular California-based script-hosting service. NeuroVerse obtained Jones’ scripts from Ubiquitous. Ubiquitous’ Terms of Use, to which Jones’ production company agreed, allegedly granted broad—but vaguely defined—rights to third parties for data “analytics and improvement of AI systems”.

The Terms of Use included in relevant part:

“By using this service you acknowledge that certain data arising from or related to your use may be accessed, analyzed, or otherwise utilized by us and authorized third-parties for operational, analytical, or technical purposes including, but not limited to, enhancing service functionality, refining user experience, and contributing to the development or improvement of automated or intelligent tools. The specific nature, scope, and duration of such data use may vary and are subject to these Terms of Use. Notwithstanding the foregoing, you retain all copyrights and ownership rights to your content, subject to any rights granted under these Terms.”

The NeuroVerse model, now licensed by GiantStreamer, a leading global streaming service, has begun autonomously generating script drafts that mirror Jones’ narrative structure, character arcs, tone, and even specific lines of dialogue.

NeuroVerse has publicly claimed that its use of copyrighted materials falls under the Fair Use Doctrine, emphasizing the transformative nature and public benefit of AI innovation. However, NeuroVerse also entered into confidential data-sharing partnerships with several Hollywood studios, further muddying the line between “scraped” data and voluntarily supplied content.

Meanwhile, the U.S. Copyright Office has pushed back against arguments that AI training constitutes a “non-expressive” use or mimics human learning processes. A related class-action lawsuit involving another generative AI company was recently allowed to proceed past the motion-to-dismiss stage, suggesting courts may be willing to hear infringement claims related to these issues.

Draft a memorandum analyzing the following legal issues:

• Copyright Infringement by NeuroVerse:
  Does NeuroVerse’s use of Jones’ original script to train its AI model constitute copyright infringement under current U.S. law?
• Defenses:
  What defenses could NeuroVerse plausibly assert against Jones’ copyright infringement claim, and what factors would weigh for or against such defenses?
• Secondary Liability and Downstream Use:
  Could Jones assert a viable claim for copyright infringement against a third-party company, such as GiantStreamer, that uses the NeuroVerse-trained AI model to generate script drafts?