Thaler v. Vidal

Thaler v. Vidal is the definitive U.S. ruling on whether AI can be named as an inventor on a patent. In August 2022, the Federal Circuit held that only natural persons can be inventors under the Patent Act, rejecting computer scientist Stephen Thaler's attempt to list his AI system DABUS as the sole inventor on two patent applications. The Supreme Court declined to hear the case in April 2023.

Background

Stephen Thaler is a computer scientist who developed an AI system called DABUS (Device for the Autonomous Bootstrapping of Unified Science). Thaler filed two patent applications with the U.S. Patent and Trademark Office naming DABUS as the sole inventor. One application was for a fractal-shaped food container, the other for a light beacon designed to attract attention during emergencies.

Thaler didn't claim he invented these items himself with AI assistance. He claimed DABUS independently conceived the inventions without any human involvement in the creative process. He listed DABUS as the inventor and himself as the assignee (the owner of the patent rights).

The USPTO rejected both applications on the straightforward ground that an inventor must be a natural person. Thaler challenged the rejection in the U.S. District Court for the Eastern District of Virginia, which upheld the USPTO's decision. He then appealed to the Federal Circuit, the specialized appellate court that handles patent cases.

What Happened

The Federal Circuit heard oral arguments and issued its opinion on August 5, 2022. Judge Leonard Stark wrote for the panel. The case turned on a question of statutory interpretation: does the Patent Act allow a non-human entity to be listed as an inventor?

Thaler argued that the Patent Act doesn't explicitly require inventors to be human and that denying AI inventorship would discourage innovation. He pointed out that DABUS autonomously generated the inventions without human direction and that someone needed to be listed as inventor to obtain patent protection.

The government, represented by the USPTO, argued that the text of the Patent Act uses language that assumes human inventors. The statute refers to "individuals" and uses personal pronouns like "himself" and "herself." These terms have a settled legal meaning: natural persons.

The Ruling

The Federal Circuit affirmed the district court and the USPTO. Judge Stark's opinion held that the Patent Act's use of "individuals" and personal pronouns unambiguously limits inventorship to natural persons. An AI system cannot be listed as an inventor on a patent application.

The court emphasized that this was a textual analysis. The statute says what it says. Congress chose the word "individuals," which the Supreme Court has consistently interpreted to mean natural persons. The personal pronouns "himself" and "herself" reinforced this reading.

Thaler petitioned the Supreme Court for certiorari. On April 24, 2023, the Supreme Court declined to hear the case, letting the Federal Circuit's ruling stand as the final word on AI inventorship in the United States. Without Congressional action to amend the Patent Act, AI systems cannot be named as inventors.

Outcome: The Federal Circuit affirmed that only natural persons can be 'inventors' under the Patent Act. The Supreme Court denied certiorari on April 24, 2023, letting the ruling stand.

Why This Case Matters

Thaler v. Vidal forced the entire patent community to restructure how it handles AI-assisted inventions. Since AI can't be an inventor, the human who "significantly contributed" to the invention must be listed. This created a framework where attorneys must identify and document the human creative input in any AI-assisted invention.

The ruling doesn't prevent patenting AI-assisted inventions. It prevents naming AI as the inventor. The practical result is that a human who uses AI as a tool in the inventive process can still be named as inventor, provided they made a significant intellectual contribution beyond simply instructing the AI to generate ideas. The line between AI-as-tool and AI-as-inventor is where patent prosecution now lives.

The case influenced similar proceedings worldwide. The UK Supreme Court reached the same conclusion in Thaler v. Comptroller-General in December 2023. The European Patent Office rejected DABUS applications. South Africa was the only major jurisdiction to initially grant a DABUS patent, and that decision was later overturned. The global consensus, with very few exceptions, is that inventors must be human.

Lessons for Attorneys

For patent attorneys, the practical takeaway is documentation. When clients use AI in their inventive process, document the human contribution at every stage. Who defined the problem? Who set the parameters? Who evaluated the AI's output? Who selected among alternatives? Who refined the concept? These human decisions are what establish inventorship, and they need to be recorded contemporaneously, not reconstructed after the fact.

For IP attorneys advising companies that use AI in R&D, build inventorship analysis into the development workflow. Don't wait until the patent application is ready to figure out who the inventor is. Identify the human contributors as the work progresses. If the AI is doing the creative work and humans are just pressing buttons, you have an inventorship problem that no amount of after-the-fact documentation can fix.

For attorneys watching the broader AI-and-law space, Thaler v. Vidal is a reminder that courts interpret statutes as written. Congress wrote the Patent Act with human inventors in mind. Until Congress changes the law, AI can't be an inventor. The same principle applies to copyright (Thaler v. Perlmutter) and will apply to other areas of IP law. If the statute assumes a human actor, AI doesn't qualify.

The Bottom Line

AI can't be an inventor under U.S. patent law. Thaler v. Vidal settled this definitively, and the Supreme Court let the ruling stand. Patent attorneys must identify and document human contributions to AI-assisted inventions or risk unpatentable innovations.