The Supreme Court leaves the door open

On March 2, 2026, the U.S. Supreme Court declined to hear the appeal, leaving the lower courts' rulings intact. This decision effectively cements the legal baseline: artificial intelligence alone cannot hold copyright. By refusing to grant certiorari, the Court allowed the U.S. Court of Appeals for the D.C. Circuit's judgment to stand, reinforcing the principle that human authorship remains a non-negotiable requirement for copyright protection.

The D.C. Circuit's ruling was explicit. It held that works created entirely by AI systems, without sufficient human creative intervention, are not eligible for copyright registration. This stance aligns with the U.S. Copyright Office's long-standing policy that only works of human authorship are protected. The Court's inaction signals a reluctance to expand copyright law to cover non-human creators, at least for now.

This legal landscape means that individuals or companies relying solely on AI to generate creative content face significant risks. Without human input that meets the threshold of authorship, the resulting works remain in the public domain. Courts have consistently emphasized that mere prompts or basic editing do not constitute sufficient human authorship. The burden remains on the human creator to demonstrate substantial, original contribution.

The implications are far-reaching. For businesses integrating AI into their creative workflows, this ruling underscores the need for robust human oversight. Copyright protection is not automatic for AI-assisted works; it must be earned through meaningful human involvement. Until the Supreme Court revisits this issue, the "human authorship rule" remains the dominant legal standard.

The Human Authorship Threshold

U.S. copyright law draws a strict line: protection applies only to works created by human beings. Because artificial intelligence lacks legal personality, it cannot hold rights or responsibilities, nor can it independently claim authorship over generated content. The Copyright Office has consistently maintained that works produced by machines without sufficient human creative input are not registrable. This principle, rooted in the Copyright Act’s requirement of human authorship, means that simply prompting an AI tool does not automatically grant copyright protection.

The critical question is not whether AI was used, but how much creative control the human exercised over the final output. Courts and the Copyright Office look for evidence of traditional elements of authorship—such as selection, coordination, and arrangement—that demonstrate the human’s independent intellectual creation. If the AI generates the work autonomously based on a general prompt, the resulting text, image, or code is generally considered public domain. Protection requires more than a request; it requires direction.

Consider the difference between two scenarios. In the first, a user types "write a sonnet about love" into an AI model and publishes the result verbatim. This output lacks the necessary human authorship because the user did not shape the specific expression. In the second scenario, a writer uses AI to generate ten draft paragraphs, then carefully edits, rewrites, and structures those paragraphs into a cohesive narrative, making deliberate choices about word choice, tone, and flow. In this second case, the human’s creative contributions—the specific expression chosen and arranged—may be eligible for copyright protection, even if the raw material originated from an AI.

The Copyright Office’s guidance emphasizes that protection extends only to the human-authored portions of a work. When AI-generated content is combined with human edits, the office conducts a "separable" analysis. It identifies which elements are the product of human creativity and which are machine-generated. Only the former is protected. This means that while the final work might be copyrighted, the underlying AI-generated fragments remain free for others to use. Authors must be able to distinguish their own creative contributions from the machine’s output to secure and enforce their rights.

When you register a work that includes AI-generated material, the U.S. Copyright Office requires you to disclose the use of AI and specify which elements were created by a human. This disclosure is not optional; it is the first step in determining whether the work meets the statutory requirement of human authorship.

The Copyright Office distinguishes between content generated entirely by AI and content where a human has made significant modifications. AI-generated text, images, or code that are presented without human intervention are not protected by copyright. However, if a human author selects, coordinates, or arranges the AI-generated material in a way that reflects original creative judgment, that specific human contribution may be protected.

To illustrate this threshold, consider the difference between uploading a prompt and editing the output. A prompt alone, such as "generate a logo for a bakery," does not constitute authorship. But if you take the AI-generated logo, then manually redraw key elements, adjust colors, and refine typography to create a unique design, the final work may contain copyrightable subject matter. The protection applies only to the human-authored modifications, not the underlying AI-generated base.

The registration process involves submitting a copy of the work along with a written statement detailing the AI use. The Copyright Office will examine this statement to determine the extent of human involvement. If the human contribution is deemed minimal or mechanical, the registration may be limited or denied. This approach ensures that copyright protection remains tied to human creativity, even in an era of advanced AI tools.

The United States has drawn a firm line: copyright protection requires human authorship. This standard, reinforced by recent Copyright Office guidance and court rulings, means that purely machine-generated output remains in the public domain. However, this strict interpretation is not the global norm. Other major jurisdictions are approaching the issue through different legal lenses, creating a fragmented international landscape for AI-generated content.

In the European Union, the debate centers on updating existing frameworks rather than establishing new authorship thresholds from scratch. The European Parliament has called for existing copyright rules to be adapted to new technologies like AI, signaling a move toward regulatory clarity rather than immediate prohibition. This approach suggests that while the EU may not grant copyright to AI itself, it is actively considering how to protect the human investment and curation involved in using these tools.

This divergence creates significant uncertainty for global businesses. A work that might be protected in one jurisdiction due to specific human editorial contributions could remain unprotected in another. Companies must now navigate a patchwork of standards, where the definition of "human intervention" varies by region. Understanding these local nuances is essential for managing intellectual property risk in a market where the legal boundaries are still being defined.

Protecting your AI-assisted IP strategy

The legal landscape for AI-generated content remains defined by a single, rigid requirement: human authorship. As of 2026, the United States Copyright Office and federal courts continue to deny copyright protection to works created entirely by artificial intelligence. This means that raw outputs from generative models are effectively in the public domain, offering no exclusive rights to the user who prompted them.

To secure intellectual property, businesses must treat AI as a drafting tool rather than a creator. The threshold for protection lies in the extent of human creative intervention. Courts look for evidence of traditional authorship, such as substantial editing, curation, and structural arrangement of the AI’s output. A simple prompt is insufficient; the human must contribute original expression that transforms the machine’s raw data into a protectable work.

Documenting this process is critical. Companies should maintain clear records of the iterative editing steps taken to refine AI-generated drafts. This documentation serves as evidence of human authorship if ownership is ever challenged. Relying solely on AI outputs for core IP assets leaves your business vulnerable to infringement claims and offers no legal recourse against competitors using the same public domain material.