Why Nothing Generated By AI is Copyrightable in the U.S. (At Least Not Yet)
Under current U.S. copyright law and the Copyright Office's evolving guidance, content generated by AI without meaningful human authorship is not eligible for copyright protection. It can't be owned. It sits in the public domain the moment it's created, available for anyone to copy, reproduce, modify, or sell without permission or payment to whoever prompted it into existence.
That's a significant legal fact that most people using AI tools professionally don't know, and it has real implications for how AI-generated content can and can't be used in commercial contexts.
The foundation of U.S. copyright law is the requirement of human authorship. Copyright protection attaches to original works of authorship created by human beings. This principle predates AI entirely. Courts have consistently held that non-human entities, including animals and, in earlier cases, divine inspiration, cannot hold copyright. The monkey selfie case, in which a court ruled that a photograph taken by a macaque was not eligible for copyright because the macaque wasn't human, established a clear precedent: authorship requires a human author.
The US Copyright Office has applied this principle directly to AI-generated content. In a series of guidance documents and registration decisions beginning in 2023, the Copyright Office has taken the position that content generated autonomously by AI is not copyrightable. The office has registered works that contain AI-generated elements where a human author made sufficient creative choices in selecting, arranging, or modifying that content, but has refused registration for works where the AI made the creative decisions and the human's contribution was primarily providing a prompt.
The Thaler case made this explicit. Stephen Thaler applied for copyright registration for an image generated by an AI system he called the Creativity Machine, listing the AI as the author. The Copyright Office rejected the application. Thaler sued. The district court upheld the rejection in 2023, ruling that human authorship is a prerequisite for copyright protection under US law. As of this writing, the case has not been definitively resolved at the appellate level, but the Copyright Office's position has remained consistent throughout.
What counts as sufficient human authorship is the genuinely uncertain question, and the Copyright Office has tried to provide guidance without drawing a bright line that the law doesn't support. A human who uses AI as a tool to execute a creative vision, making specific choices about which outputs to use, how to arrange them, what to modify, and how to combine them with human-created elements, may have a copyright claim to the resulting work. A human who types a prompt and accepts the first output is probably not the author in the legal sense, because the creative choices were made by the system, not the person.
The practical implications depend on what you're using AI-generated content for. If you're generating marketing images, blog drafts, or product descriptions with AI tools and using them internally, the copyright question may not matter much. If you're building a business on AI-generated content, selling it, licensing it, or relying on copyright protection to prevent competitors from copying it, the absence of copyright protection is a serious gap. Anyone can take your AI-generated content and use it freely, because it belongs to everyone and no one simultaneously.
There's a further complication on the other side of the transaction. While AI-generated outputs may not be copyrightable, the training data used to create AI systems very often is. Lawsuits against AI companies from artists, writers, photographers, and news organizations allege that training on copyrighted works without permission constitutes infringement. These cases are working their way through the courts and have not been definitively resolved. The legal status of AI training on copyrighted data is genuinely unsettled in a way that the copyright status of AI outputs is, at least for now, somewhat more settled.
Other jurisdictions are taking different approaches. The UK has a provision in its copyright law that extends protection to computer-generated works, attributing authorship to the person who made the arrangements necessary for the work to be created. The European Union is developing its own framework under the EU AI Act that may treat the question differently. China has had court decisions going in various directions. The international picture is fragmented in ways that create complications for content used across borders.
The phrase "at least right now" in this piece's title is doing real work. Copyright law evolves through legislation and court decisions, and the current US framework was written long before generative AI existed. Congress has the authority to extend copyright protection to AI-generated works if it chooses to, and there are arguments on both sides of that policy question. Lobbying from content industries, technology companies, and creator communities is ongoing. The legal landscape in five years may look substantially different from today's.
For now, the practical guidance is straightforward: if copyright protection matters for your use case, AI-generated content in the US doesn't have it unless a human author made sufficient creative contributions to the final work. Document those contributions if you're trying to establish a copyright claim. And if you're building a product or business that depends on owning the content you create, understand that AI-generated content currently provides no such ownership, and plan accordingly.