In 2018 and 2019, Dr. Stephen Thaler filed a patent application naming Device for Autonomous Bootstrapping of Unified Sentience, known as “DABUS,” as the inventor of the resulting inventions. The application was filed in the European Union, United States, and United Kingdom patent offices, and all three entities denied the patent application on the basis of one key point: Only human inventors can be issued a patent.
In the United States, the Patent Act provides the statutory support for the conclusion that inventors must be humans. Section 101 of the Patent Act provides, “Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor.…” Section 102 adds that a “person shall be entitled.…”
In its Decision on Petition, in which it rejected the patent application naming DABUS as an inventor, the U.S. Patent and Trademark Office (“USPTO”) explained that “conception—the touchstone of inventorship—must be performed by a natural person.”
While the USPTO’s decision was based on an existing bright-line legal rule (i.e., human or not?), the robust policy debate around AI and patents requires speculation about how the law can and should (or should not) evolve to address AI’s role in inventorship.
It’s important to note that the really hard questions do not necessarily involve inventions that contain some aspect of AI technology. The patent world has been grappling with those issues for decades. According to the USPTO, “in the 16 years from 2002 to 2018, annual AI patent applications increased by more than 100%, rising from 30,000 to more than 60,000 annually. Over the same period, the share of all patent applications that contain AI grew from 9% to nearly 16%.”
What is new and novel is that we are now faced with AI that, on its own, can invent. Indeed, AI continues to evolve from a mere tool that facilitates the creation of innovative outputs to creating outputs itself—a painter not just a paintbrush. In 1950, Alan Turing posed the question: “Can machines think?” We must now also consider: “Can machines invent?”
The team behind the Artificial Inventor Project, the petitioners pushing for DABUS to be recognized as an inventor, certainly think so. In an article published in the June, 2019 edition of the World Intellectual Property Organization (“WIPO”) magazine, Ryan Abbott of the Artificial Inventor Project argues that not recognizing AI as an inventor could stifle innovation and lead to misaligned incentives.
“It is important that appropriate policies are put in place to deal with AI-generated works. Today, inventive AI may be a relatively insignificant part of innovation in economic terms. But AI is improving exponentially, and human researchers are not. Even in the short-to-medium term, this means that inventive AI may become a significant part of research and development. When it does, it will be seriously problematic if we lack clear rules on whether AI-generated inventions can be protected, who, or what, should be listed as an inventor, and who owns these inventions and related patents.”
The Pressing Policy Questions Concerning AI and Patents
It’s impossible to know what’s coming next for AI and patent law policy. As Yogi Berra once said, “It's tough to make predictions, especially about the future.” The only thing we know for sure is that machines will continue to become more advanced and technological innovation will continue to accelerate. The challenge for those who will shape IP law in the years to come—legislators, regulators, judges and lawyers, to name a few—is finding a middle ground that fosters a partnership between machines and humans that furthers fundamental patent law policy objectives of promoting innovation and investment in new technologies.
The debate is well underway. In 2019, WIPO held the “First Session of the WIPO Conversation on IP and AI” to discuss the impact of Al on IP policy, with the objective being to gather feedback from the global IP community on the “questions that policymakers need to ask.”
Among other things, this effort resulted in the publication of the “Revised Issues Paper on Intellectual Property Policy and Artificial Intelligence.” It’s important to note that, while this article is focused exclusively on issues related to AI and patents, AI impacts many areas of IP, and the Issues Paper takes that breadth and scope into account. In addition to patents, it addresses copyright, data, designs, trademarks, trade secrets, technology gap and capacity building, accountability for IP administrative decisions, and feedback regarding commonly used terms (such as “AI” and “AI-generated”) in an effort to land on agreed definitions.
The patent issues, and related questions, identified in the paper are categorized as follows:
Inventorship and Ownership
Representative Questions: (1) Should the law require that a human being be named as the inventor or should the law permit an AI application to be named as the inventor? (2) If AI-generated inventions are excluded from patent protection, what alternative protection mechanisms are available for such inventions?
Patentable Subject Matter and Patentability Guidelines
Representative Questions: (1) Should AI-generated and AI-assisted inventions be treated in the same way as other computer-implemented inventions? (2) If AI applications or algorithms fall within exclusions from patentability would that incentivize AI applications and algorithms to be kept as secrets and exacerbate the so-called black box problem?
Inventive Step or Non-Obviousness
Representative Questions: (1) What implications will having an AI application replacing a person skilled in the art have on the determination of the prior art base? (2) Should AI-generated content qualify as prior art?
Representative Questions: (1) How should data used to train an algorithm be treated for the purposes of disclosure? (2) Should the human expertise used to select data and to train the algorithm be required to be disclosed?
General Policy Considerations for the Patent System
Representative Questions: (1) Should AI-generated inventions benefit from patent protection? (2) If so, would it be enough to incorporate AI-generated inventions in the current legal system or should consideration be given to a sui generis system of IP rights for such inventions in order to adjust incentives for AI?
These are just a few of the issues—related only to patents—that the paper raises. Clearly there are many different and difficult legal, economic and philosophical questions that must be sorted through in the years to come. Right now we have more questions than answers.
Notably, the patent section of the paper concludes with the following:
Is it too early to consider these questions because the impact of AI on both science and technology is still unfolding at a rapid rate and there is, at this stage, insufficient understanding of that impact or of what policy measures, if any, might be appropriate in the circumstances?
While many of the policy questions posed above involve lots of complexity and uncertainty, the question of whether it’s too early to be considering these issues seems to have a more straightforward answer. Given the pace of technological advancement, and the economic rewards to be reaped by securing valuable patents, it’s all but certain that AI will continue to play an increasingly important role in inventorship. Our global patent system, therefore, needs to be prepared, and a big part of the preparation involves grappling with the difficult policy questions we are confronted with.
If you're interested in talking more about AI as it relates to your IP, contact Ben Stasa at email@example.com. Want to learn more? Check out this other article by Ben: Should AI Systems Be Eligible for Patents?