Recently, artificial intelligence (AI) has become an increasing part of our daily lives. Many of us utilize virtual assistants such as Apple’s Siri and Amazon’s Alexa, we interact with customer support through chat bots, and we receive individualized content curated by Netflix, Facebook, and Spotify. These services are increasingly powered by artificial intelligence, which some define as any artificial system that can perform tasks under varying and unpredictable circumstances, without significant human oversight, or that can learn from their experience and improve their performance. Such intelligent machines can include computing systems that apply machine learning or deep learning architectures that mimic the neural networks of the human brain.
Although the term artificial intelligence and the first AI programs were developed in the 1950s, the development of AI-based systems is now accelerating at a pace surpassing even experts’ expectations. This strongly suggests that in the near future, AI-enabled systems will become even more deeply engrained in our society. Within a few years, self-driving cars, AI-powered health care (e.g., diagnosis, personalized drug development, and treatment), and personal assistants (that can do much more than merely set timers and control household appliances) will be commonplace. Additionally, improvements in artificial general intelligence (AGI) systems may eventually challenge human intelligence in terms of creativity and inventive talent. With this context, intellectual property practitioners should reexamine the fundamental aspects of IP creation when artificially-intelligent systems are involved.
This article explores how current trademark, copyright, and patent law may apply to scenarios in which an AI, without significant human oversight, might place goods and services in commerce, “create” works, or “invent” novel and non-obvious innovations.
AI and Trademark Law
Trademark law serves multiple purposes including to: a) assure a potential customer that goods or services with a distinguishing mark are made by the same producer as other similarly marked goods or services; and b) assure a producer that an imitating competitor will not be able to benefit from financial or reputation-related rewards associated with a desirable good or service. The Lanham Act accomplishes these purposes by allowing a person who is an owner of a trademark used in commerce to register that trademark on a Federal principal register. Under the Act’s definitions, the term “person” need not include a “natural person,” but could also include “a juristic person” such as a legal entity “capable of suing and being sued in a court of law.”
Today, AI-based services predict purchases, recommend products to customers, and place specific products in commerce through online shopping. Assuming the services act substantially without human intervention and are placing goods and services into the stream of commerce (e.g., by offering them for sale in an advertisement), the relevant inquiry is whether a trademark right could inure from such AI behavior, and if so, to what entity would the right inure.
Here, if the AI is considered a juristic person, it could apply for, and eventually obtain the trademark. If the AI is not a juristic person for the purposes of the Lanham Act, it could not itself apply for a trademark. However, the Act does not require the specific entity that places the goods or services in commerce to apply for the trademark. Accordingly, under current law, an operator or an owner of the AI could apply for trademarks on behalf of the AI-based service.
AI and Copyright Law
The “IP Clause” of the United States Constitution grants Congress the power to secure for “authors” an exclusive right to their respective works. The Copyright Act later codified the Constitution to provide protections for “original” works of “authorship” fixed in any tangible medium of expression. However, neither the Constitution nor the Copyright Act defines “author” or authorship.
For the purposes of copyright, Supreme Court jurisprudence has defined an author as “the party who actually creates the work, that is, the person who translates an idea into a fixed, tangible expression entitled to copyright protection.” Recently, this arguably ambiguous definition of author has been tested by advocates for non-human authors. For example in Naruto v. Slater, legal representatives for a photo-taking crested macaque asked a district court to recognize the animal as author (and rightful owner) of copyrighted photographs (dubbed the Monkey Selfies). In its analysis, the court stated that the Copyright Act does not have any “mention of animals anywhere . . . . [and that] the Supreme Court and Ninth Circuit have repeatedly referred to ‘persons’ or ‘human beings’ when analyzing authorship under the Act.” Without deciding the merits of authorship or ownership, the district court dismissed, stating that the Copyright Act does not confer standing to non-human animals to sue for copyright infringement.
By rule, the United States Copyright Office enforces a “Human Authorship Requirement” and “will refuse to register a claim if it determines that a human being did not create the work.” Furthermore, the Copyright Office will not register works produced by “a machine or mere mechanical process that operates randomly or automatically without any creative input or intervention from a human author.”
Today, AI-based applications are regularly creating works without significant human oversight that would normally qualify for copyright protection. For example, the Associated Press has been using AI to write and publish articles. Furthermore, an AI bot created an entire screenplay, which was subsequently made into a short sci-fi film called Sunspring.
Such real-world examples raise the issue of whether an original and fixed work created solely by AI could obtain copyright protection. However, current Copyright Office rules and copyright jurisprudence indicate that solely-AI-created works cannot obtain copyright protections.
In 1979, shortly after the most recent Copyright Act, a National Commission concluded that there was “no reasonable basis for considering that a computer in any way contributes authorship to a work produced through its use.” However, in 1986, a Congressional Advisory Panel questioned such a restrictive interpretation of computer operations:
It is still an open question whether the programmed computer is unlike other tools of creation. Authorities in the field of artificial intelligence (AI), although disagreeing on AI nature and purpose, do agree that its aim is to produce a pattern of output that would be considered intelligent if it were displayed by a human being. One must ask, therefore, whether machines or interactions with machines might produce a pattern of output that would be considered creative or original if done by a human being. If machines are in any sense co-creators, the rights of programmers and users of programs may not be easily determined within the present copyright system.
For this reason, some argue that AI-based applications should be treated under the work made for hire doctrine, which considers a corporate entity as a legal author of a work for which it is not the author-in-fact. Other jurisdictions such as Hong Kong, India, Ireland, New Zealand, and United Kingdom have interpreted or amended their copyright laws under similar doctrinal theories. For example, in the UK, the author of computer-generated artistic works is assumed to be “the person by whom the arrangements necessary for the creation of the work are undertaken.”
AI and Patent Law
In addition to copyright, the IP Clause of the Constitution provides Congress the power to secure to inventors a limited exclusive right to their respective discoveries – a patent. Like “authors,” the Constitution does not define “inventors.” However, “inventor” is defined in the Leahy–Smith America Invents Act (AIA) as “the individual or, if a joint invention, the individuals collectively who invented or discovered the subject matter of the invention.” The AIA also requires that “a person shall be entitled to a patent unless . . .,” suggesting that only a person can be an inventor. Additionally, pre-AIA law requires that a person contribute to the conception of an invention. Accordingly, patent law requires an inventor to be a person.
AI-based applications are currently developing products ranging from car components to drug molecules that would normally qualify for patent protection. One example is the cross-bristle design for the Oral-B Cross Action toothbrush created by the Creativity Machine, an AI-based construct.
Such real-world examples raise the issue of whether a novel invention created solely by AI can obtain patent protection. Under current patent law, AI-based systems cannot be “inventors” because they are not “persons.” Furthermore, if no human can be said to have been involved in, or conceived of, any element of any claim, such a patent would be seemingly invalid for lacking an inventor, as defined by the AIA.
Today, artificial intelligence-based computing systems are routinely outperforming humans in a variety of narrow tasks due to advances in fields such as machine learning and neural networks. These breakthroughs should cause practitioners, and society at large, to seriously consider whether solely-AI-created works and inventions should garner IP protection. While an AI “operator” or owner can currently seek federal rights under trademark law, existing US copyright and patent laws require a human creator of a copyright work and a human inventor for each patent claim. As more generalized forms of artificial intelligence begin to match and surpass human cognitive abilities, it will be interesting to see whether and how Congress might expand intellectual property law to accommodate AI-authorship and invention.
 National Security Commission Artificial Intelligence Act of 2018, H.R. 5356, 115th Cong. § 2(i)(1) (2018).
 Patrick Caughill, AI is Developing Faster than Experts Imagined. Do We Need a Speed Limit?, Futurism (Sep. 1, 2017), https://futurism.com/ai-is-developing-faster-than-experts-imagined-do-we-need-a-speed-limit/.
 See U.S. Food and Drug Administration, FDA Permits Marketing of Artificial Intelligence-Based Device to Detect Certain Diabetes-Related Eye Problems (Apr. 11, 2018), https://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm604357.htm.
 “General intelligence is what people do” and what AGI seeks to accomplish. Kate Baggaley, There Are Two Kinds of AI, and the Difference is Important, Popular Science (Feb. 23, 2017), https://www.popsci.com/narrow-and-general-ai#page-2.
 See Qualitex Co. v. Jacobson Prod. Co., 514 U.S. 159, 163 (1995).
 15 U.S.C. §§ 1051(a)(1); 1051(a)(3); 1051(a)(3)(C) (2002).
 15 U.S.C. § 1127 (2006).
 See, e.g., James Vincent, A Bot on Amazon is Making the Best, Worst Smartphones Cases, The Verge (Jul. 10, 2017), https://www.theverge.com/tldr/2017/7/10/15946296/amazon-bot-smartphone-cases; Amit Sharma, How Predictive AI Will Change Shopping, Harvard Business Review (Nov. 18, 2016), https://hbr.org/2016/11/how-predictive-ai-will-change-shopping.
 15 U.S.C. § 1051(a)(1).
 U.S. Const. art. I, § 8, cl. 8.
 17 U.S.C. § 102(a) (1990).
 Cmty. for Creative Non-Violence v. Reid, 490 U.S. 730, 737 (1989).
 Naruto v. Slater, No. 15-CV-04324-WHO, 2016 WL 362231, at *1 (N.D. Cal. Jan. 28, 2016), aff’d, No. 16-15469, 2018 WL 1902414 (9th Cir. Apr. 23, 2018).
 Id at *3.
 Id. at *4.
 Compendium of U.S. Copyright Office Practices, § 306 (3d ed.).
 Id. at § 313.2.
 Ross Miller, AP’s ‘Robot Journalists’ Are Writing Their Own Stories Now, The Verge (Jan. 29, 2015), https://www.theverge.com/2015/1/29/7939067/ap-journalism-automation-robots-financial-reporting.
 HAL 90210, This is What Happens When an AI-Written Screenplay is Made into a Ailm, The Guardian (Jun. 10, 2016), https://www.theguardian.com/technology/2016/jun/10/artificial-intelligence-screenplay-sunspring-silicon-valley-thomas-middleditch-ai; Amanda Kooser, AI-Written Film ‘Sunspring’ a Surreal Delight, Upchucked Eyeball Included, CNET (Jun. 13, 2016), https://www.cnet.com/news/ai-written-film-sunspring-a-surreal-delight-upchucked-eyeball-included/ (“‘Sunspring’ is chaotic, super weird and still worth watching.”).
 National Commission on New Technological Uses of Copyrighted Works (CONTU), Final Report, 109 (1979).
 U.S. Cong., Office of Technology Assessment, Intellectual Property Rights in an Age of Electronics and Information, 72 (1986).
 17 U.S.C. § 201(b) (1978).
 Andrews Guadamuz, Artificial Intelligence and Copyright, WIPO Magazine (Oct. 2017), http://www.wipo.int/wipo_magazine/en/2017/05/article_0003.html.
 UK Copyright Law, § 9(3) of the Copyright, Designs and Patents Act (CDPA) (1988).
 Constitution, supra note 10.
 35 U.S.C. § 100(f) (2015).
 35 U.S.C. § 102(a) (2015) (emphasis added).
 “The threshold question in determining inventorship is who conceived the invention. Unless a person contributes to the conception of the invention, he is not an inventor.” Manual of Patent Examining Procedure (MPEP) § 2137.01 (Rev. Jul. 2015).
 Daniel Terdiman, Inside The Hack Rod, The World’s First AI-Designed Car, Fast Company (Dec. 1, 2015), https://www.fastcompany.com/3054028/inside-the-hack-rod-the-worlds-first-ai-designed-car; Rachel Brazil, Artificial Intelligence: Will it Change the Way Drugs are Discovered?, The Pharmaceutical Journal (Dec. 7, 2017), https://www.pharmaceutical-journal.com/news-and-analysis/features/artificial-intelligence-will-it-change-the-way-drugs-are-discovered/20204085.article.
 Peter Rejcek, When the Mother of Invention Is a Machine, Who Gets Credit?, Singularity Hub (Nov. 3, 2016), https://singularityhub.com/2016/11/03/when-the-mother-of-invention-is-a-machine-who-gets-credit/#sm.001umdagggs9dtl1001297df69zj8.
© 2018 McDonnell Boehnen Hulbert & Berghoff LLP
snippets is a trademark of McDonnell Boehnen Hulbert & Berghoff LLP. All rights reserved. The information contained in this newsletter reflects the understanding and opinions of the author(s) and is provided to you for informational purposes only. It is not intended to and does not represent legal advice. MBHB LLP does not intend to create an attorney–client relationship by providing this information to you. The information in this publication is not a substitute for obtaining legal advice from an attorney licensed in your particular state. snippets may be considered attorney advertising in some states.