the firm's post-grant practitioners are some of the most experienced in the country.

Steven B. Chang
David M. Longo, Ph.D.
Eric W. Schweibenz
Teddy S. Gron
Richard D. Kelly
John F. Presper
Jay E. Rowe, Jr., Ph.D.
Jianping (James)  Wu
Tia D. Fenton
Alec M. Royka
Erik M. Stang, Ph.D.
Norman F. Oblon
Kevin Ross  Davis
Maki  Saitoh
Robert T. Pous
Andrew M. Ollis
Christopher I. Donahue
J. Derek  Mason, Ph.D., CLP
Peifang  Tian, Ph.D.
James R. Love
Nanlin  Wang, Ph.D.
Diane  Jones
Anna Z. Lloyd
Yuki  Onoe
Alexander B. Englehart
Vincent K. Shier, Ph.D.
John  Sipos
Grace E. Kim
Johnny  Ma
Daniel J. Pereira, Ph.D.
Ryan W. Smith
Charles L. Gholz
Kevin L. Hartman, Ph.D.
Sameer  Gokhale
Kasumi  Kanetaka
Kurt M. Berger, Ph.D.
Aristotelis M. Psitos
Yanwen  Fei
Chika (Teranishi) Iitoyo
Craig R. Feinberg
Yin Y. Nelson, Ph.D.
Robert W. Downs
Tao  Feng, Ph.D.
Stefan Uwe  Koschmieder, Ph.D.
Robert  Tarcu
Long  Phan, Ph.D.
John S. Kern
Nicholas  Rosa, Ph.D.
Colin B. Harris
Soumya  Panda
Yorikatsu  Hohokabe, Ph.D.
Jeffrey B. McIntyre
Akihiro  Yamazaki
Brian B. Darville
Kevin M. McKinley
Ching-Cheng (Tony)  Chang
Philippe J.C. Signore, Ph.D.
Frank J. West
Stephen G. Baxter, Ph.D.
Surinder  Sachar
Carl E. Schlier
Derek  Lightner, Ph.D.
Thomas M. Cunningham, Ph.D.
Bogdan A. Zinchenko
Arthur I. Neustadt
Marina I. Miller, Ph.D.
Dale M. Shaw
Elissa L. Sanford
Eckhard H. Kuesters
Edwin D. Garlepp
Michael R. Casey, Ph.D.
Christopher  Ricciuti
Jenchieh (Joseph) Yuan
Aldo  Martinez
Yuanyi (Alex) Zhang, Ph.D.

Technologies

Artificial Intelligence (AI)
Artificial Intelligence (AI)
Digital Health
Digital Health
Energy & Renewables
Energy & Renewables

Fast Facts

About Our

Law Firm

About Our Law Firm

Headquartered within steps of the USPTO with an affiliate office in Tokyo, Oblon is one of the largest law firms in the United States focused exclusively on intellectual property law.

Get to know our

History

Get to know our History

1968
Norman Oblon with Stanley Fisher and Marvin Spivak launched what was to become Oblon, McClelland, Maier & Neustadt, LLP, one of the nation's leading full-service intellectual property law firms.

Our Local and

Global Reach

Our Local and Global Reach

Outside the US, we service companies based in Japan, France, Germany, Italy, Saudi Arabia, and farther corners of the world. Our culturally aware attorneys speak many languages, including Japanese, French, German, Mandarin, Korean, Russian, Arabic, Farsi, Chinese.

A few of our

ACCOLADES

A few of our ACCOLADES

Oblon's professionals provide industry-leading IP legal services to many of the world's most admired innovators and brands.

OPPORTUNITIES FOR YOUR

Career

OPPORTUNITIES FOR YOUR Career

From the minute you walk through our doors, you'll become a valuable part of a team that fosters a culture of innovation, client service and collegiality.

A few ways to

GET In Touch

A few ways to GET In Touch
US Office

Telephone: 703-413-3000
Learn More +


Tokyo Office

Telephone: +81-3-6212-0550
Learn More +

Downloadable

Patent Forms

Downloadable Patent Forms

The United States Patent and Trademark Office (USPTO) issued final rules implementing the inventor's oath or declaration provisions of the America Invents Act (AIA) on August 14, 2012.

Update on DABUS: Can an AI Machine be an Inventor?

  • October 18, 2021
  • Article

Associated People

Associated Technologies


As previously discussed in this blog (see Creation vs. Conception: Can an AI Machines be an Inventor?) some AI researchers believe that the output of certain sophisticated AI software can represent a new invention conceived of, not by the author, user, or trainer of the software, but by the software itself, and that the software or the “AI machine” should be listed as the inventor on a patent application for the new invention. This AI-machine-as-inventor proposition is currently being tested by Dr. Stephen Thaler via the filing of patent applications in over a dozen countries for inventions allegedly invented by a neural-network-based AI machine (“DABUS”), and the listing of the inventor as “DABUS” on the application (with Dr. Thaler listed as the applicant and assignee).  At the time of our previous posts last fall, the USPTO, the EPO, and the UKIPO each objected to listing DABUS as the inventor since DABUS is not a natural person.

Briefly reviewing, according to Dr. Thaler, DABUS includes “…a first artificial neural network, made up of a series of smaller neural networks, that has been trained with general information from various knowledge domains…A second ‘critic’ artificial neural network monitors the first neural network for new ideas, and identifies those ideas that are sufficiently novel compared to the machine’s pre-existing knowledge base.” Ryan Abbott, artificialinventor.com (2020).  Further, Dr. Thaler argues that DABUS was not created to solve any particular problem, but only received training in general knowledge in a field.  Moreover, Dr. Thaler contends that DABUS itself “…proceeded to independently conceive of the invention, and to identify it as novel and salient.”  Id.

In particular, the filed patent applications asserted to have been invented by DABUS are directed to a “Neural Flame” (a warning light for attracting enhanced attention, including a controllable light source configured to pulse based on an input signal of a lacunar pulse train having specific fractal properties) and a “Fractal Container” (a food or beverage container having a generally cylindrical wall with a fractal profile of convex and concave fractal elements). 

Since our previous post, courts in four countries have weighed in on whether DABUS can be listed as the inventor in a patent application based on those inventgions.

U.S. and U.K. Cases

On August 6, 2020, Dr. Thaler filed a complaint against the USPTO in the Eastern District of Virginia (Stephen Thaler v. Andrei Iancu and the USPTO) asking for declarative and injunctive relief that would set aside the USPTO decision.  The complaint asserts that “requiring a natural person to be listed as an inventor as a condition of patentability is contrary to law,” and that the USPTO actions were arbitrary, capricious, and an abuse of discretion, unsupported by substantial evidence.

On September 2, 2021, the judge in that case ruled against Thaler, asserted that under the plain language of the Patent Act, an AI machine cannot be an inventor.  In particular, echoing the rationale set forth in the USPTO decision, the judge noted that U.S. law consistently refers to an inventor as an “individual,” which various courts have interpreted to mean a “natural person.”  Thus, the judge suggested that Thaler’s best hope would be a change in U.S. law, stating that “it will be up to Congress to decide how, if at all, it wants to expand the scope of patent law.”

Also in September, an appellate court in the U.K. upheld the UKIPO’s decision to reject Thaler’s U.K. application, because, under U.K. law, a “person” must be named as the inventor.

South African and Australian cases

In contrast, in July, the South African patent office granted two patents designating DABUS as the inventor, primarily based on the fact that the World Intellectual Property Organization (WIPO) accepted that designation for the corresponding PCT applications.  In each case, Dr. Thaler was listed as the applicant.

Also in July, an Australian federal court ruled that inventions invented by an AI machine can qualify for patent protection, reversing a decision of the Australian patent office that only a person could be listed as an inventor.  In particular, the court considered the question of whether or not an AI machine could be listed as an inventor on a PCT application, which merely requires listing the “name of the inventor.”

In its analysis, the court first noted that nothing in Australian patent laws and regulations specifically excludes AI machines or systems from being listed as the inventor.

Next, the court noted that since the term “inventor” is not defined in the Australian patent law, the ordinary meaning of the term should be used.  In particular, the court stated than the word “inventor” is an agent noun, like “controller” and “regulator,” and describes the agent that does the act referred to by the verb to which the “er” or “or” is attached, and that such an agent can be a person or a thing.  Thus, under this reasoning, if an AI machine is an agent that invents, it can be the “inventor.”

Third, the court argued that it would be inaccurate not to list the AI machine as the inventor since, in many cases, a person cannot be said to have been the inventor of the claimed invention. In this regard, the court discussed the multiple networks employed by DABUS, including a second neural work that perceives inventive value in the output of first network, i.e., recognizes the invention.  Thus, the court stated that DABUS can “mimic aspects of human brain function” and is not just a “brute force computational tool.” 

The court further stated that the question of ownership and control of an invention (restricted to legal persons) is a completely separate question from inventorship, and acknowledged that DABUS could be neither an applicant for a patent nor a grantee of a patent under Australian law.  The court also acknowledged that it was not clear whether Dr. Thaler would ultimately qualify as both applicant and grantee if DABUS was listed as the inventor (e.g., how does he obtain title), but stated that the Australian patent office need not decide that question at this time, but should determine only whether or not a valid PCT application was filed.

Finally, the court noted that Australian patent law is more concerned with “inventive step” and not the mental state or actions of an inventor.  Accordingly, the court appears to dismiss any suggestion by the patent office of a “consciousness” requirement for inventors, reasoning that only the inventive step, however it arose, is important.

Clearly pleased with the South African and Australian decisions, Dr. Thaler’s team asserts that these rulings will encourage companies to develop AI technology in those particular countries that provide such protection for AI machines.  Going forward, it will be interesting to see whether other countries will follow the lead of South Africa and Australia, and whether Congress will consider changes to U.S. patent law to allow for AI inventorship.