Published online by Cambridge University Press: 15 February 2019
The field of robotics is remarkably wide, with many social settings now entailing and increasingly requiring the use of robots to support a variety of human activities. Unsurprisingly, robots’ form and shape, their level of intelligence and intended purpose can vary significantly depending on the relevant industry.1 Domestic robots are already a reality in a growing number of family homes. They include both humanoid robots which support those in need (such as the elderly, people with disabilities or children) and robots for household consumer markets, including domestic vacuum cleaners and grass-trimmers. Humanoid robots only account for a small fraction of the industry, with robotic arms for industrial automation being instead widespread.2
Dr Enrico Bonadio and Dr Luke McDonagh are Senior Lecturers in intellectual property law at The City Law School, City, University of London. They are both part of a group of researchers that have been awarded funding by the EU as part of Horizon2020 to assess the area of interactive robots in society (INBOTS project). Christopher Arvidsson is a Graduate Teaching Assistant and PhD candidate at The City Law School, City, University of London, and conducts research on the law and emerging technologies.
1 See discussion by H James Wilson, “What Is a Robot, Anyway?” (2015) Harvard Business Review, available at <hbr.org/2015/04/what-is-a-robot-anyway>.
2 CA Keisner, J Raffo, S Wunsch-Vincent, “Breakthrough technologies – Robotics, innovation and intellectual property”, Economic Research Working Paper No 30 (WIPO Publication) at p 4.
3 See ISO 8373:2012(en) “Robots and robotic devices – Vocabulary”, available at <www.iso.org/obp/ui/#iso:std:iso:8373:ed-2:v1:en>. ISO is an international organisation with a membership of 161 national standards bodies. Through its members, ISO develops voluntary International Standards which aims to support innovation and provide solutions to global challenges (www.iso.org/about-us.html). ISO 8373 is an International Standard that specifies vocabulary used in relation to inter alia robots operating in industrial and non-industrial environments. As such, ISO 8373 provides definitions and explanations of the most commonly used terms that are grouped into clauses by main topics of robotics. See also IFR, Service Robots - Definition and Classification WR 2016, available at <ifr.org/img/office/Service_Robots_2016_Chapter_1_2.pdf>. The IFR was established as a non-profit organisation in 1987 and consists of members from the robotics industry, national or international industry associations and research & development institutes. Amongst other things, the organisation provides market data, stimulates research in robotics and promotes links between science and industry (ifr.org/association).
4 See <www.iso.org/obp/ui/#iso:std:iso:8373:ed-2:v1:en>; IFR (Service Robots) at <ifr.org/img/office/Service_Robots_2016_Chapter_1_2.pdf>; IFR (Industrial Robots) at <ifr.org/img/office/Industrial_Robots_2016_Chapter_1_2.pdf>.
6 See Executive Summary at <ifr.org/downloads/press/Executive_Summary_WR_Service_Robots_2017_1.pdf>, p 12.
7 ibid, p 14.
8 Although exploring a comprehensive definition of AI is beyond the scope of this article, it is thought of as an area of technology that focuses on the creation of intelligent machines that work and react like human beings. Examples of technologies that can currently be termed AI include spam filters, algorithmic trading systems and medical systems generating diagnoses (see <ifr.org/downloads/papers/Media_Backgrounder_on_Artificial_Intelligence_in_Robotics_May_2018.pdf>, p 1).
9 Keisner, Raffo, Wunsch-Vincent, supra, note 2, at p 7.
10 See IFR Artificial Intelligence in Robotics, May 2018 at <ifr.org/downloads/papers/Media_Backgrounder_on_Artificial_Intelligence_in_Robotics_May_2018.pdf>, p 1.
11 ibid, p 1.
12 See the European Parliament resolution of 16 February 2017 with recommendations to the Commission on Civil Law Rules on Robotics (2015/2103(INL)), letter P.
13 See the discussion in Keisner, Raffo, Wunsch-Vincent, supra, note 2, at p 12.
14 See for example C Andrew Keisner, J Raffo and S Wunsch-Vincent, “Breakthrough Technologies – Robotics and IP”, December 2016, Economics and Statistics Division, WIPO 2016, available at <www.wipo.int/wipo_magazine/en/2016/06/article_0002.html>.
15 See “Making Your Robotics Company a More Attractive Investment” (2012) Robotic Business Review, 21 October 2012, at <www.roboticsbusinessreview.com/unmanned/making_your_robotics_company_a_more_attractive_investment>.
18 See Keisner, Raffo, Wunsch-Vincent, supra, note 2, at p 27.
19 The European Patent Office is based in Munich (Germany), its activity, and the patents it grants, being regulated by the European Patent Convention (Convention on the Grant of European Patents (European Patent Convention: “EPC”) of 5 October 1973 as revised by the Act revising Article 63 EPC of 17 December 1991 and the Act revising the EPC of 29 November 2000).
20 LJ Thayer, RL Emsley, “Be Competitive: Patent Planning for Robotics Companies” (2011) Robot Magazine, available at <www.finnegan.com/en/insights/be-competitive-patent-planning-for-robotics-companies.html>.
21 See the webpages of Finnegan law firm at <www.finnegan.com/images/content/8/6/v3/866/IntellectualPropertyConsiderationsfortheRoboticsIndustry-revised.pdf>.
22 iRobot Corporation v Urus Industrial Corporation Case 1:05-cv-10914-RGS.
24 ibid (Final Consent Judgment); see also iRobot’s announcement of the settlement at <investor.irobot.com/news-releases/news-release-details/irobot-settles-patent-and-copyright-infringement-lawsuit>.
25 ibid (Final Consent Judgment).
26 iRobot Corporation v Robotic FX, INC, Civil Action No CV-07-RRA-1511-S.
27 iRobot Corporation v Robotic FX INC, CA No 2:07-cv-1511-RRA (Stipulated Consent Judgement and Permanent Injunction).
29 The patents in question are EP 1 331 537 B1, EP 2 251 757 B1, EP 1 969 438 B1, and EP 1 395 888 B1.
30 See iRobot’s press release at <investor.irobot.com/news-releases/news-release-details/irobots-preliminary-injunctions-against-shenzhen-silver-star>.
31 The aim of this piece of legislation is to harmonise national regimes on the protection of confidential information.
32 Keisner, Raffo, Wunsch-Vincent, supra, note 2, at p 27.
35 See discussion at <www.finnegan.com/images/content/8/6/v3/866/IntellectualPropertyConsiderationsfortheRoboticsIndustry-revised.pdf>, p 3.
36 Keisner, Raffo, Wunsch-Vincent, supra, note 2, at p 32.
37 See discussion at <www.finnegan.com/images/content/8/6/v3/866/IntellectualPropertyConsiderationsfortheRoboticsIndustry-revised.pdf>, p 3.
39 Alice Corp v CLS Bank International 573 US, 134 S Ct 2347 (2014).
40 A couple of reported disputes focusing on alleged misappropriation of robot-related industrial secrets have occurred in the US. We have already mentioned iRobot’s lawsuit against Robotic FX Inc in the US District Court for the District of Northern Alabama. It is interesting to note that iRobot filed a separate lawsuit against the same defendant before a court in Massachusetts for violation of trade secrets. The Massachusetts Court ruled that the defendant had misappropriated iRobot’s proprietary and confidential information, violated fair trade practices and destroyed evidence: see iRobot’s press release at <investor.irobot.com/news-releases/news-release-details/irobot-prevails-lawsuits-against-robotic-fx>. Another relevant legal action was started in 2013 by Mako Surgical against Blue Belt Technologies and Mako’s former sales manager who had joined the defendant for a position working on the Navio surgical system. The defendants had been sued for alleged trade secret misappropriation and for the former employee’s violation of the non-compete obligation to his former employer (Mako Surgical). Mako obtained an order preventing that employee (former Mako’s sales manager) from working for BBT for a certain period, in any position, and Blue Belt Technologies from using confidential information: see the article “MAKO Surgical Sues Blue Belt Technologies”, in Robotic Business Review of 10 June 2014 at <www.roboticsbusinessreview.com/legal/mako_surgical_sues_blue_belt_technologies>.
41 See Council Directive 91/250/EEC of 14 May 1991 on the legal protection of computer programs, repealed by Directive 2009/24/EC.
42 See also C Leroux et al, “EU Robotics Coordination Action: A green paper on legal issues in robotics” (2012) p 29, available at <www.unipv-lawtech.eu/files/euRobotics-legal-issues-in-robotics-DRAFT_6j6ryjyp.pdf>.
43 Alleged copyright infringement has also been the focus of the above mentioned dispute iRobot v Urus Industrial Corporation (which was eventually settled). That case was not just about alleged infringement of the patents covering certain functional aspects of the Roomba. iRobot also claimed that Urus’ vacuum cleaner robot infringed copyright in iRobot’s product literature and system interface, including its musical audio feedback features.
44 Keisner, Raffo, Wunsch-Vincent, supra, note 2, at p 34.
45 Art 6 of Directive 2001/29. A similar provision is set forth by the US Digital Millennium Copyright Act (DMCA): a rule which was invoked in 2001 by Sony when a programmer created and distributed via a website free software to enhance the capabilities of the robot dog named Aibo produced by said Japanese company (such user basically decrypted the code defining the robotic dog’s abilities). Sony complained that the website in question provided the means to circumvent the copy protection protocol of Sony’s AIBO memory stick to allow access to the relevant software, and therefore constituted a violation of the DMCA anti-circumvention provision. On this case see M Rimmer, “Respect the Code or the Dog Gets It” (December 2001) InCite, p 31, available at <eprints.qut.edu.au/98504/1/aibo.pdf>.
51 See M Chakravorty et al, “Design-Patent Protection for Modern Robotics Companies: What to Do When the Face of Your Robot Becomes the ‘Face’ of Your Company” (2014) Robotics Business Review, July 1, p 6.
52 EU Regulation 6/2002.
56 Keisner, Raffo, Wunsch-Vincent, supra, note 2, at p 31.
63 Keisner, Raffo, Wunsch-Vincent, supra, note 2, at p 31.
65 See, amongst others, Ramalho, A, “Will Robots Rule the (Artistic) World? A Proposed Model for the Legal Status of Creations by Artificial Intelligence Systems” (2017) 21(1) Journal of Internet Law 12 Google Scholar ; Guadamuz, A, “Do androids dream of electric copyright? Comparative analysis of originality in artificial intelligence generated works” (2017) 2 Intell Prop Q 169 Google Scholar ; Abbott, R, “Artificial intelligence, big data and intellectual property: protecting computer-generated works in the United Kingdom” in T Aplin (ed), Research Handbook on Intellectual Property and Digital Technologies (Edward Elgar 2017)Google Scholar ; Denicola, RC, “Ex Machina: Copyright Protection for Computer-Generated Works” (2016) 69 Rutgers L Rev 251 Google Scholar ; McCutcheon, J, “The Vanishing Author in Computer-Generated Works: A Critical Analysis of Recent Australian Case Law” (2013) Melbourne University Law Review 36 Google Scholar ; Bridy, A, “Coding Creativity: Copyright and the Artificially Intelligent Author” (2012) Stan Tech L Rev 5 Google Scholar ; Abbott, R, “I Think, Therefore I Invent: Creative Computers and the Future of Patent Law” (2016) 57(4) Boston College Law ReviewGoogle Scholar ; Hattenbach, B and Glucoft, J, “Patents in an Era of Infinite Monkeys and Artificial Intelligence” (2015) 19 Stanford Technology Law Review 32 Google Scholar .
66 For the sake of clarity, we will use in the remainder of the article the terms “machine”, “robot” and “computer” interchangeably.
71 Artworks produced by AI have also started being auctioned: see for example a sale by Christie of an AI-created portrait in a gilt frame in October 2018 (see <www.christies.com/features/A-collaboration-between-two-artists-one-human-one-a-machine-9332-1.aspx>).
72 Copyright, Designs and Patents Act 1988, s 178.
73 Abbott (2017), supra, note 65, p 2.
74 Abbott (2017), supra, note 65, p 3.
75 Express Newspapers plc v Liverpool Daily Post & Echo plc and Others  1 WLR 1089.
77 de Cock Buning, M, “Autonomous Intelligent Systems as Creative Agents under the EU framework for Intellectual Property” (2016) European Journal of Risk Regulation 311 Google Scholar .
78 See <www.thepaintingfool.com>; Simon Colton, a Professor of Computational Creativity in the Department of Computing of Goldsmiths College, University of London, is the academic behind the Painting Fool.
82 See also Bridy, supra, note 65, p 4 (discussing the generative art movement, which aims at exploring “computational creativity” via “a set of creative practices whereby the artist cedes control to a system that is self-contained enough to operate autonomously”).
83 The fixation requirement, provided by several copyright acts (for example, in the US and UK), will not be analysed here. There is indeed little doubt that most machine-produced works meet this requirement.
84 Case C-5/08 Infopaq International A/S v Danske Dagblades Forening at .
85 See respectively Art 1(3) of the Software Directive (Directive 2009/24), Art 6 of the Copyright Term Directive (Directive 93/98) and Art 3(1) of the Database Directive (Directive 96/9).
86 Case C-145/10 Eva-Maria Painer v Standard Verlags GmbH at .
87 Case C-145/10 Eva-Maria Painer v Standard VerlagsGmbH at . See also Case C-604/10 Football Dataco at  (citing also, by analogy, Case C-393/09 Bezpecnostni Softwarova Asociace at  and , and Joined Cases C-403/08 and C-429/08 Football Association Premier League and Others at ).
88 Case C-145/10 Eva-Maria Painer v Standard VerlagsGmbH at .
89 de Cock Buning, supra, note 77, p 314.
90 Case C-145/10 Eva-Maria Painer v Standard VerlagsGmbH, Opinion of Advocate General Trsteniak, 12 April 2011, at .
93 Case C-145/10 Eva-Maria Painer v Standard VerlagsGmbH and Others at –.
94 Most copyright acts, including in the US and UK, however, still do not provide a definition of “authors”. Only judicial decisions address what authorship means, and who an author is (see JC Ginsburg, “The Concept of Authorship in Comparative Copyright Law” (2002) 52 DePaul L Rev 1063 at 1066). In the US for example the Supreme Court has defined an author as “he to whom anything owes its origin; originator; maker; one who completes a work of science or literature” (Burrow-Giles Lithographic Co v Sarony, 111 US 53, 58 (1884)). US lower courts have also noted that an author must be more than one who contributes creativity or originality to a work (Aalmuhammed v Lee, 202 F 3d 1227, 1233 (9th Cir 2000)); basically, it must be one “who superintended the whole work, the ‘master mind’” (ibid, citing Burrow-Giles Lithographic Co 111 US at 53, 61).
95 See also Ramalho, supra, note 65, p 10.
96 See Preambulo, Ley 22/11 sobre la Propiedad Intelectual de 1987: “los derechos que corresponden al autor, que es quien realiza la tarea puramente humana y personal de creacion de la obra y que, por lo mismo, constituyen el nucleo esencial del objeto de la presente Ley”.
97 The French Code of intellectual property defines protectable subject matter as “oeuvres de l’esprit”: see Art L112-1 of the French Code de la Proprieté Intellectuelle.
98 US Copyright Office, Compendium of US Copyright Office Practices § 101 (3rd edn 2017) p 306. Examples of works that fail the authorship requirement include, for instance, a photograph taken by a monkey and works “... produced by a machine ... that operates randomly or automatically without any creative input or intervention from a human author” (see Compendium 313.2).
99 Case C-145/10 Eva-Maria Painer v Standard VerlagsGmbH, Opinion of Advocate General Trsteniak, 12 April 2011, at 121.
100 Guadamuz, supra, note 65, p 178.
101 Guadamuz, supra, note 65, p 178.
102 That authorship is a requirement under US copyright law is confirmed by a string of cases that have dealt with who should be considered author in situations where “celestial voices” may have played a role (psychography cases): Oliver v Saint Germain Foundation 41 F Supp 296 (SD Cal 1941); Urantia Foundation v Burton, No K 75-255 CA 4, 1980; Urantia Foundation v Maaherra 895 F Supp 1337 (D Ariz 1995); Penguin Books, Inc v New Christian Church of Full Endeavour No 96 CIV 4126. Indeed, when faced with claims of supernational authorship, US courts in those disputes concluded that only humans can own the copyright. Similarities can also be drawn with claims of “animal authorship”, where US courts have confirmed this line of argument: see Naruto v Slater 2018 US App Lexis 10129 (9th Cir Cal, Apr 23, 2018) (rejecting the argument brought by the plaintiff, an animal rights organisation, that US copyright law does not prohibit an animal – in the case at issue, a monkey that had taken a selfie – from owning a copyright. The court held that while the animal had constitutional standing it “lacked statutory standing to claim copyright infringement of photographs)”.
103 Guadamuz, supra, note 65, p 179.
104 Namely, Ireland, New Zealand, India, South Africa and Hong Kong.
105 See again the Copyright, Designs and Patents Act 1988, s 178.
106 McCutcheon, supra, note 65, at pp 44–45.
107 Guadamuz, supra, note 65, p 176.
108 For an opposite opinion see Lionel Bently, mentioned by B González Otero and JP Quintais, “Before the Singularity: Copyright and the Challenges of Artificial Intelligence” (2018) Kluwer Copyright Blog, available at <copyrightblog.kluweriplaw.com/2018/09/25/singularity-copyright-challenges-artificial-intelligence/> (stressing that the CDPA provisions on computer-generated works do not offer a useful model for protecting AI outputs, because of their incompatibility with the EU copyright acquis and failure to address the issue of originality).
109 Nova Productions v Mazooma Games  EWCA Civ. 219.
110 ibid, at .
111 Guadamuz, supra, note 65, p 177.
112 This scenario may soon become a common reality, for example in the field of computer games, taking into consideration the increasingly relevant roles of users in such interactive games: see Ramalho, supra, note 65, p 11.
113 See also Denicola, supra, note 65, pp 283–285 (noting that, should users be considered the authors of computer-generated works, there would no longer be a need to distinguish between machine-generated works and machine-aided works. Indeed, if the creative output is considered to be authored by someone other than the user of the machine (or is not deemed copyrightable at all), it would be necessary to differentiate situations where the computer is simply a tool from scenarios where the machine itself is the creator. Denicola thus believes that the “users’ option” should be preferred. The market – he stresses – already supplies programmers with the incentive to create software and machines which are used to create works, such incentive mainly coming from the prospect of maximising sales. Also, should programmers be deemed as authors and owners of the copyright over the work produced by the machine – his argument goes – users may not be encouraged to purchase and use the software to create works as they would not be the copyright holders: which would eventually jeopardise the interests of programmers, who may see sales of their machines decrease. Finally, attributing copyright to programmers may turn out to be useless, as programmers themselves would frequently be unaware of the creation of works by users.
114 Khouri, AH, “Intellectual Property Rights for Robots: on the Legal Implications of Human-like Robots as Innovators and Creators” (2017) 35(3) Cardozo Arts & Entertainment Law JournalGoogle Scholar .
115 Khouri, supra, note 114, p 668.
116 See also McCutcheon, supra, note 65, p 952 (noting that “while the incentive of copyright may be irrelevant to a novelist compelled to write their opus, it may well explain why an expensive computer-generated production is made. Without that reward, the work may not be made ... or disseminated”).
117 See Tatiana Synodinou, mentioned by González Otero and Quintais, supra, note 108.
118 Ramalho, supra, note 65, pp 16–20.
119 Database Directive (Directive 96/9).
120 Art 4 of Copyright Term Directive (Directive 93/98) provides that “Any person who, after the expiry of copyright protection, for the first time lawfully publishes or lawfully communicates to the public a previously unpublished work, shall benefit from a protection equivalent to the economic rights of the author. The term of protection of such rights shall be 25 years from the time when the work was first lawfully published or lawfully communicated to the public”.
121 Hattenbach and Glucoft, supra, note 65, p 35 (reporting that said brute-force computing has also been used to create defensive publications aimed at preventing others from patenting in the same field).
122 Hattenbach and Glucoft, supra, note 65, pp 36 and 51 et sqq (noting however that many of these claims appear nonsensical, and that therefore their validity needs to be assessed on a case-by-case basis).
123 See US patent No 5,659,666.
124 Abbott (2016), supra, note 65, pp 1083–1086.
125 See US patent No 5,852,815.
126 Abbott (2016), supra, note 65, p 1081.
127 EPC Rule 19(1); see also Art 41(2)(j) of the EPC confirming that the request for a grant of a European patent shall contain the designation of the inventor, where the applicant is the inventor.
128 EPC Rule 19(1).
129 Abbott (2016), supra, note 65, p 7.
130 European IPR Helpdesk, “Fact Sheet Inventorship, Authorship and Ownership” March 2013 see <www.iprhelpdesk.eu/sites/default/files/newsdocuments/Fact-Sheet-Inventorship-Authorship-Ownership.pdf>, at p 3.
132 PCT stands for Patent Cooperation Treaty, an international treaty administered by the World Intellectual Property Organization (WIPO) which assists applicants in obtaining patent protection internationally for their inventions and helps offices, including the EPO, with their decisions to grant or refuse the patent.
134 Yet, it has also been noted that certain mechanically-generated claims may be considered obvious. For example, the linguistic manipulation software devised by the company Cloem (see above, including note 121) often merely adds or deletes sentences. As noted by Hattenbach and Glucoft, supra, note 65, p 45, many of these claims are “the result of relatively slight rearrangements, and these minor modifications that work in predictable ways would by definition be considered obvious”.
136 Abbott (2016), supra, note 65, p 1091.
137 Abbott (2016), supra, note 65, pp 1124–1125.
138 Abbott (2016), supra, note 65, p 1125.
139 Abbott (2016), supra, note 65, p 1114.
140 See “The IP Battle Continues for Robotics Companies”, in Robotic Business Review of 7 January 2018, at <www.roboticsbusinessreview.com/legal/the_ip_battle_continues_for_robotics_companies>.
141 See also Maggiore, M, “Artificial Intelligence, Computer-Generated Works and Copyright” in E Bonadio and N Lucchi (eds), Non-Conventional Copyright – Do New and Atypical Works Deserve Protection? (Edward Elgar 2018) p 391 Google Scholar .