On March 2, 2026, the United States Supreme Court declined to take up the question of whether works generated entirely by artificial intelligence are eligible for copyright protection. The Court’s denial of certiorari effectively closes the book, at least for now, on a years-long legal saga brought by Stephen Thaler, a Missouri-based computer scientist who waged a persistent campaign to secure intellectual property rights for the output of his AI systems. This denial of certiorari underlines the Copyright Act’s requirement of human authorship for copyright protection but just how much human authorship is required, in the fast-evolving world of AI, remains unanswered.
Background
The case traces back to 2018, when Thaler sought to register “A Recent Entrance to Paradise,” a piece of visual art with the U.S. Copyright Office, naming an AI system he developed (DABUS) as the work’s creator. Thaler acknowledged that the image was created by AI without traditional human authorship but claimed that the copyright should be extended because he owned and programmed the AI system. The Copyright Office rejected the application in 2022, finding that copyrightable creative works require human authors.
Thaler challenged that determination in federal court. In 2023, US District Court Judge Beryl A. Howell ruled squarely against him, holding that “human authorship is a bedrock requirement of copyright.” The U.S. District Court for the District of Columbia upheld that ruling in 2025. Thaler then petitioned the Supreme Court for review, contending that the lower courts’ decisions had “created a chilling effect on anyone else considering using AI creatively.” The Supreme Court’s refusal to hear the case leaves the appellate ruling intact.
The Requirement of Human Authorship Has Been Consistently Reinforced
This outcome is part of a broader and consistent stance in copyright and patent law, advanced in large part by Thaler himself. Thaler filed two patent applications listing DABUS as the sole inventor. Both were refused by the USPTO, that concluded the Patent Act limits inventorship to natural persons. The U.S. Court of Appeals for the Federal Circuit affirmed that decision in Thaler v. Vidal, 43 F.4th 1207 (Fed. Cir. 2022) and held that AI systems cannot be listed as inventors on patents because they are not natural persons. The USPTO reinforced that position in its Revised Inventorship Guidance for AI-Assisted Invention published in November 2025, clarifying that while AI cannot itself be named as an inventor, human inventors may still use AI-powered tools in the inventive process and obtain patent protection for the resulting work. Likewise, in January 2025 the Copyright Office issued its report on Copyright and Artificial Intelligence, Part 2: Copyrightability affirming that “copyright protection in the United States requires human authorship.”
Implications for IP Practitioners and Creators
The Supreme Court’s decision not to intervene leaves creators and their attorneys with a clear, if still evolving, framework. Output generated purely by AI, created without meaningful human creative contribution, remains outside the scope of copyright protection in the United States. And just how much human contribution is required? Well, the bar appears high. In two recent U.S. Copyright Decisions, Zarya of the Dawn and Theatre D’opera Spatial, protection was denied to AI-generated images despite human creators inputting a high number of detailed prompts. In Zarya of the Dawn, the Copyright Office approved protection for the text and arrangement of images but denied protection for the individual images within the graphic novel. Copyright protection was denied because the Office deemed the AI tool Midjourney the creator of the images’ “expressive elements,” despite the user inputting thousands of prompts to create those images. Similarly, in Theatre D’opera Spatial, AI artist Jason M. Allen, used more than 624 detailed, iterative prompts to create an award-winning image “Theatre D’opera Spatial” with the AI system Midjourney. The Copyright Office, in denying his copyright application, found that Allen did not exercise sufficient control over Midjourney’s final output because the AI itself determined the final image’s structure and tone. Allen has requested that the U.S. District Court for the District of Colorado overturn the Copyright Office’s refusal to register his work and that case is still pending. See Allen v. Perlmutter, D.Colo. 1:24-cv-02665.
Despite these recent holdings, many creators are using AI tools to generate art, and that trend is only increasing, as entertainment behemoths themselves are investing in AI companies. So, how do creators using AI tools protect themselves? Without clear guidance as to how much human intervention is necessary for AI-generated works to be copyrightable, best practices include placing human creativity at the very core of the creative process and documenting all instances of human intervention.
