The question of who – or what – can be an author has become one of the most contentious issues in modern copyright law. As artificial intelligence systems generate increasingly sophisticated creative works, courts and policymakers are grappling with a fundamental question: Does copyright law require a human author?
The Traditional Framework
Copyright law has long operated on an implicit assumption: authors are human. The U.S. Copyright Act protects “original works of authorship,” but nowhere does it explicitly define what constitutes an “author.” For most of copyright’s history, this omission didn’t matter. Authorship was self-evident because humans were the only entities capable of creating expressive works.1
The requirement that authors be human isn’t merely traditional – it is rooted in copyright’s constitutional purpose. The U.S. Constitution grants Congress the power to secure exclusive rights to authors in order “to promote the Progress of Science and useful Arts.”2 This incentive structure presumes rational actors who can be motivated by the promise of exclusive rights. Can a machine be incentivized? Does an AI system need copyright protection to continue creating?
The Monkey Selfie and Compendium Guidance
The “human” requirement entered the public discourse with Naruto v. Slater,3 the famous “monkey selfie” case (discussed earlier on the blog). In 2011, a crested macaque in Indonesia grabbed photographer David Slater’s camera and snapped several selfies. When Slater sought to enforce copyright in these images, the People for the Ethical Treatment of Animals (PETA) sued on behalf of the monkey, arguing the animal should own the copyright.4
The courts rejected this claim, and the U.S. Copyright Office updated its examination guidelines to make explicit what had been implicit: “The Office will register an original work of authorship, provided that the work was created by a human being.”5 The Compendium clarifies that works created by animals, plants, or divine beings cannot be registered. Crucially, it also states that works “produced by a machine or mere mechanical process” without creative human input are uncopyrightable.6
AI-Generated Works and Recent Decisions
The rise of generative AI has forced courts to apply these principles in new contexts. In Thaler v. Perlmutter,7 Dr. Stephen Thaler sought to register an image created entirely by his AI system, “Creativity Machine,” listing the AI as the author. The D.C. Court of Appeals affirmed the Copyright Office’s refusal to register,8 holding that copyright requires human authorship. (Our analysis was discussed earlier on the blog.)
The court reasoned that copyright’s purpose is to incentivize human creativity. If AI systems can create without legal protection, there’s no policy justification for extending copyright to their outputs. The decision emphasized that “copyright law only extends to works created by humans” and that this requirement is consistent with centuries of precedent.
The Spectrum of Human Involvement
The more nuanced question isn’t whether fully autonomous AI creations deserve protection, but rather: How much human involvement is necessary to secure copyright?
Current guidance suggests that works created with AI assistance may be copyrightable if there’s sufficient human creative control. The Copyright Office has indicated it will register works that reflect “the author’s own original mental conception, to which [they] gave visible form.” Using AI as a tool—similar to using Photoshop or a camera—doesn’t preclude copyright protection if the human author makes creative choices about prompts, selection, arrangement, or modifications.
However, simply entering a text prompt into an AI system may not suffice. The Copyright Office has suggested that minimal prompting resulting in unpredictable outputs may not demonstrate sufficient human authorship. The key distinction appears to be whether the human exercised meaningful creative control over the expressive elements of the final work.
Looking Forward
The human authorship requirement raises profound questions about the future of creative production. As AI systems become more sophisticated, the line between tool and creator blurs. Should copyright law adapt to protect AI-generated works to incentivize investment in AI technology? Or does maintaining the human authorship requirement serve important policy goals about the nature and purpose of creative expression? Thaler has recently petitioned for writ of certiorari with the U.S. Supreme Court, appealing the D.C. Court of Appeals decision.
Some scholars argue that recognizing AI authorship could lead to a flood of low-value copyrights that would actually hinder creativity rather than promote it. Others contend that significant human investment in training and operating AI systems deserves legal protection.
For now, the law remains clear: copyright requires a human author. But as technology continues to evolve, so too will the debates about who, or what, deserves ownership of creative works. The fundamental question remains not just legal but philosophical: What makes something worthy of copyright protection, and what role should humanity play in that calculus?
This blog post provides general information about copyright law and should not be construed as legal advice. Copyright law is complex and fact-specific; consult with a qualified attorney for guidance on specific situations. For more information, please contact Yonaxis I.P. Law Group.
- See 17 U.S.C. §102(a). ↩︎
- U.S. Const., Art. I, Sec. 8, Cl. 8. ↩︎
- 888 F.3d 418 (9th Cir. 2018). ↩︎
- The Naruto case was decided on its merits, but it was technically dismissed for lack of “next friend” standing by PETA to sue on behalf of an animal, namely, Naruto. ↩︎
- Compendium of the U.S. Copyright Office Practices, 3rd ed. (Jan. 2021) (“Compendium 3rd”), §306; see also Burrow-Giles Lithographic Co., 111 U.S. 53, 58 (1884). ↩︎
- Compendium 3rd, §313.2. ↩︎
- 687 F. Supp. 3d 140 (D.D.C. 2023). ↩︎
- 130 F.4th 1039 (D.C. Cir. 2025). ↩︎