Undergraduate Law Review at Florida State University

Thaler v. Perlmutter and the Inherent Humanity of Art and Media

—

by

Undergraduate Law Review at FSU
in

Writer: Dylan Rivas

Editor: Aidan Larkin

Associate Editor: Jake Taylor

Artwork Note: The artwork featured in this post is the AI-generated work at issue in Thaler v. Perlmutter. We have included the artwork so that readers can view the work discussed in this article and consider for themselves the questions it raises about creativity, authorship, and the role of artificial intelligence in the creation of art and media.

I. Introduction

    On March 2, 2026, The United States Supreme Court denied a writ of certiorari by computer scientist Dr. Stephen Thaler that argued for the copyright protection of a painting-esque artwork generated by an artificial intelligence (AI) system Thaler had created.1 In doing so, the Court left in place past rulings that cited numerous ownership provisions outlined by the Copyright Act of 1976 that specifically reference the human aspect of registration, including the assumption that a work’s protection is based on the author’s lifetime and hinges on their signature.2 The Court declined to review Thaler’s arguments for artificial intelligence as a copyright-approved author, but the extent to which work that involves AI is protected by the American legal system remains unclear.3 The current review system allows an author to claim copyright on portions of their work that do not involve AI. However, this system is flawed in its reliance on highly situational criteria for human ingenuity. Thus, instead of merely classifying AI as a tool for creativity, a broader rejection of AI-aided content’s copyright protection is necessary.4

    II. Case Law and Precedent

      The practice of necessitating human authorship within the United States Copyright Office has been in place long before the development of modern artificial intelligence. Routinely referred to as the Intellectual Property Clause, Article I, Section 8, Clause 8 of the United States Constitution specifically grants “to authors … the exclusive right to their writings,” which has been interpreted by the Copyright Act of 1976 to specifically apply to original works of authorship.5 However, since some of the Act’s precedent dates as far back as the 19th century, a frequent point of contention has been the relationship between the aid of technology and a user’s intention when creating a piece of intellectual property. In Burrow-Giles Lithographic Company v. Sarony (1884), general factors of authorship concerning photography were established. These factors included “facts of originality, intellectual production, and of thought and conception.”6 In this case, Napoleon Sarony, a photographer, argued that a photograph has a high level of human authorship, as the user of a camera has discretion over elements of its composition, such as scenery and camera function.7 This type of terminology continues to outline the idea of active authorship for the U.S. Copyright Office and courts to uphold.

      III. Gaps for Further Argumentation  

        Central to the U.S. Supreme Court’s denial of Thaler’s appeal was the fact that he did not, in this case, claim human authorship whatsoever. Instead, he sought protection for media created by an entity that he owns.8 This appeal also relied on the assumption that the Court would believe AI is capable of being a sentient author, an idea unaligned with the Court’s current precedent.9 However, a greater concern regarding this case stems from the practicality of the U.S. Copyright Office’s policy. The Office’s 2023 Registration Guidance allows authors to gain protection for works that constitute sufficient human action in the development process itself, claiming the parts of a work that include only their own human contribution are protected.10 A parallel case, Allen v. Perlmutter (2024), disputes the denial of a work generated by AI, arguing that the 600 prompts which the author gave to the AI model was sufficient direction and arrangement for approval.11 

        IV. The Question of Medium

          While it is apt to classify AI as a tool, the specific reasoning that a human person must be “putting pen to paper” in bringing about a work leaves contextually sound arguments against AI vulnerable to the objection that other tools and means of technology do the work for the author as well. AI similarly follows direction throughout the creative process in many cases. An AI software can have its settings manipulated in the same way as a camera—it can alter images at the click of a button in the same way an editing suite can. While surely not all uses of AI can be reduced to just this, the point is that many mediums at times can act independently of their users’ abilities, solely requiring their direction.12 In likening AI use to other devices, one might argue that only the parts of a project that are uninfluenced by those tools should be protected as intellectual property. Under this reasoning, it is difficult to distinguish between a creative who clicks to apply a brush or filter in an editing software and an individual who writes a prompt telling an AI model to apply the same brush or filter. One could even make the case that the discretionary prompt writing constitutes a higher degree of creativity. The underlying idea of this argument is that while copyright protects the intuition expressed with the tool, the work itself is judged as a whole when deciding whether it is to be protected. As this class of technology inevitably evolves, maintaining the current legal framework would muddle the process by which an author reports which elements of their work are to be considered entirely their own. This, in turn, leads to the question of what is protected under copyright law, and how it is supposed to be protected, becoming evermore unclear.

          V. Recommendation

            While Thaler v. Perlmutter (2023) did not reach the U.S. Supreme Court for further review, the implications of its ruling in the D.C. court play an integral part in a broad dialogue that highlights a need to pivot from piecemeal protections and vague standards of approval for granting copyright for creative works when a creative work employs the use of AI. The United States District Court for the District of Columbia’s decision sets an intuitive precedent that states an AI model cannot itself be classified as a copyright-protected author. While this is a logical decision, the copyright protections of work that is aided by this type of technology remain unclear. Lawmakers cannot claim that these models are conventional tools that do not serve as decision makers in the creative process, while simultaneously claiming that artists cannot use them for a project in its entirety. Because AI models can serve as decision-makers in the creative process, their involvement in a work presents a different consideration than the use of conventional creative tools. Thus, to uphold a consistent standard and acknowledge the fluidity of the interaction between person and machine, authors of works that include the direct influence of artificial intelligence cannot be offered the same copyright protection as those who create without the aid of artificial intelligence.

            1. Edward D. Lanquist et al., Supreme Court Denies Certiorari in Thaler v. Perlmutter:
              AI Cannot Be an Author Under the Copyright Act, Baker Donelson (Mar. 2026),
              https://www.bakerdonelson.com/supreme-court-denies-certiorari-in-thaler-v-perlmutter-ai-cannot-be-an-author-under-the-copyright-act (on file with the Undergraduate Law Review at FSU). ↩︎
            2. Id. ↩︎
            3.  Sasha S. Rao et al., IP Hot Topic: Cert. Denied, But One Thing is Cert-ain: The Human Authorship Requirement for Copyright, Sterne Kessler (Mar. 2026), https://www.sternekessler.com/news-insights/client-alerts/ip-hot-topic-cert-denied-but-one-thing-is-cert-ain-the-human-authorship-requirement-for-copyright/ (on file with the Undergraduate Law Review at FSU). ↩︎
            4. Christopher T. Zirpoli, Cong. Rsch. Serv., LSB10922, Generative Artificial Intelligence and Copyright Law (2025). ↩︎
            5. 17 U.S.C. § 102 (2024). ↩︎
            6. Burrow-Giles Lithographic Co. v. Sarony, 111 U.S. 53 (1884). ↩︎
            7. Id. ↩︎
            8. Thaler v. Perlmutter, No. 25-449, 714 F. Supp. 3 103, 2023 WL 5333236 (D.D.C. Aug. 18, 2023). ↩︎
            9. Scott Bomboy, Supreme Court denies artificial intelligence authorship claim for artwork copyright, Nat’l Const. Ctr. (Mar. 2026), https://constitutioncenter.org/blog/supreme-court-denies-artificial-intelligence-authorship-claim-for-artwork-copyright (on file with the Undergraduate Law Review at FSU). ↩︎
            10. Copyright Registration Guidance: Works Containing Material Generated by Artificial Intelligence, 88 Fed. Reg. 16,190, Mar. 16. 2023. ↩︎
            11.  Allen v. Perlmutter, No. 1:24-cv-2665 (D. Colo. filed 2024); Edward D. Lanquist et al., Supreme Court Denies Certiorari in Thaler v. Perlmutter: AI Cannot Be an Author Under the Copyright Act, Baker Donelson, (Mar. 2026), https://www.bakerdonelson.com/supreme-court-denies-certiorari-in-thaler-v-perlmutter-ai-cannot-be-an-author-under-the-copyright-act (on file with the Undergraduate Law Review at FSU). ↩︎
            12. Christopher T. Zirpoli, Cong. Rsch. Serv., LSB10922, Generative Artificial Intelligence and Copyright Law (2025). ↩︎

            Comments

            Leave a comment