a cura di Ana Paula Gonzalez Torres
AI-generated inventions: (A)I, inventor?
Considering the advances in artificial intelligence and the current ever changing legal framework in such regards this article examine the issue of authorship and artificial intelligence generated inventions. The initial aim is to clarify the terminology, it then proceeds to examine particular instances of artificial intelligence generated inventions and a judicial case involving the attribution of inventorship. The final goal of this article is to reflect upon the legislative changes that will be needed in light of the ever more frequent uses of artificial intelligence in the invention process from a comparative perspective.
It is important to acknowledge that, currently, we are far from the scenario depicted in “I, Robot.” Some have predicted that human-level Artificial Intelligence will be achieved around 2029, while others guess it will take another 180 years. Nonetheless, given the rising interest in “AI-generated inventions,” it is pivotal to question and clarify how, if at all, to vest inventorship to AI systems and understand the difference between AI-aided inventions and AI-autonomously generated inventions.
2. Definition of AI
“Artificial Intelligence” (AI) will be considered a branch of computer science that relies on performing mathematical methods and algorithms, thereby computer implementation. As a system, it “learns” from data and structured data by means of machine learning or operates by means of “search-based problem solving and constraint satisfaction, knowledge-based reasoning, planning, probabilistic reasoning, [or] natural language processing.”
“AI-generated inventions” will be considered as an umbrella term comprising AI-aided inventions and AI-autonomously generated inventions. Such, because both involve designing algorithm(s), training algorithm(s) on data, use of a database to run the AI system, automatization of a process, and will result in an invention.
“AI-aided inventions” will be considered as the situation where a natural person conceives an invention and uses an AI system as a tool for automatizing specific phase(s) of the invention process. The natural person explicitly designs the algorithm(s), modifies weights, structures data, runs the algorithm on the data, and obtains automated expected results.
“AI-autonomously generated inventions” will be considered as the situation in which an AI system develops an invention independently from human conception. The AI system is initially programmed by a natural person. The AI system then proceeds to perform computation through not explicit codification employing self-improving software, which utilizes mathematical random search methods or statistical methods for correlation of input data and output data, thereby achieving an unexpected result(s) that meet the requirements of patentability.
3. AI inventor?
In the past, there have been several instances of proclaimed inventions autonomously generated by AI. For example, the “Creative Machine” by Stephen Thaler, which, as early as 1994, used an artificial neural network with self-stimulation and another network that focuses on the value of the output to create new patterns of information and not just associating patterns. The “Creative Machine” has then been credited with the “creation” of numerous inventions that have been granted patents like the “Neural network based prototyping system and method.” In the field of Genetic Programming, the “Invention Machine” by John Koza has “created” patentable inventions like the “Apparatus for Improved General-Purpose PID and non-PID Controllers” with minimum human intervention by being fed basic information about components and specifications about the desired outcome. More recently, the “Device for the autonomous generation of useful information” (DABUS) by Stephen Thaler, a connectionist artificial intelligence, has been credited with the “creation” of a “Neural Flame” and a “Food Container.” By means of neural networks, the AI system is said to generate ideas and autonomously identify when those auto-produced ideas are novel compared to the machine’s pre-provided database. Such a system has allegedly autonomously generated inventions, “creations,” and led to patent applications that purposely presented as the inventor, the AI system DABUS.
The allegedly autonomously generated inventions, “creations,” of the Creative Machine and the Invention Machine were granted patents listing as the patent’s inventors their respective “programmers,” meaning the designers of the AI system, Thaler and Koza. On the other hand, DABUS, also known as the Artificial Inventor “creations,” leads to patent applications that purposely indicated as the inventor of the AI system, DABUS.
Decisions from the USA, UK, and Europe found that the applications meet patentability requirements but have denied the possibility of granting an invention to an AI system. The USPTO did not take an instance on whether DABUS “created” the inventions. Nonetheless, the USPTO rejected the applications because the relevant status uses pronouns specific to natural persons such as “himself” and “herself.” Thus, because case law considers inventorship a mental act, the invention is limited to natural persons. The EPO avoided deciding whether DABUS “created” the inventions. Instead, the EPO stated that because AI systems cannot have the rights that come from being an inventor, they have no legal personality. Thus, it did not meet the formal requirements. Meanwhile, the UKIPO accepted that DABUS created the invention but denied the applications on the grounds that legislation required the inventor to be a natural person. On the other hand, the Federal Court of Australia found that artificial intelligence is capable of being an “inventor” for the purposes of the Australian patent regime. It stands its decision on the basis that there is “no specific provision that expressly refuted the position that an artificial intelligence system can be an inventor.” Lastly, South Africa’s patent office granted the first patent to DABUS for its food container.
4. Why is this relevant?
The international scheme set up by the Patent Cooperation Treaty, the relevancy of the decision by the South Africa Patent Office, and the different views between courts in developed countries, it is thus important to understand the dimensionality of the issue at hand. For starters, case law in the USA states that “[o]ne who merely suggests an idea of a result to be accomplished, rather than means of accomplishing it, is not [an] inventor.” Similarly, in the realm of AI systems, there is a subdivision between the task of problem definition and the task of establishing the means of accomplishing a certain goal. Usually, both tasks are performed by the programmer, but recently, because of self-improving software, achieving a certain goal is not “explicitly” programmed but derived based on correlation rules. Usually, the means of accomplishing a certain goal are clearly established in an algorithm’s design, the sequence of steps that transform given inputs into the intended output, and the rules of correlation are based on statistical methods. Nonetheless, it has been said that ‘the role of human inventors in the Artificial Invention Age [will be] to formulate high-level descriptions of the problem to be solved, not to work out the details of the solution […] Once given this problem description, the artificial invention software produces a design for a concrete product […] that solves the stated problem.’).
When the programmer has provided the means to accomplish the intended goal, the programmer is still the inventor, and the AI system has been used to produce an AI-aided invention. While, when AI systems can perform tasks without being explicitly programmed by leaning on mathematical and statical methods and the “automated model fitting” has come up with an output that has vastly diverged from the original input by the programmer. Then it stops being within the “means of accomplishing it” initially set by the programmer. It thus begs the question, is it the situation in which “one who merely suggests an idea of a result to be accomplished rather than means of accomplishing it, is not [an] inventor”?
Because we are a long way from strong AI, but because of the role of computation improvements in AI, there is a need for a middle ground legal framework. It has been stated that grating patent protection to AI-generated inventions may reduce human incentives to create new inventions. Thus, it is important to acknowledge transparently and explicitly the various degrees of involvement of AI systems in the invention process. Only in such a way, the AI system could validate or incentivize, instead of diminishing, human inventions. In this regard, a useful framework is analyzing the spectrum of human intervention to distinguish between human-inventions, AI-aided inventions, and autonomous generation of inventions by AI. All of which should be deserving of patents, with a differentiated legal treatment. Otherwise, refusing to issue patents would encourage resourcing to trade secrets as protection deterring transparency goals.
Furthermore, failing to recognize differentiated legal treatment between human-inventions, AI-aided inventions, and autonomous generation of inventions by AI systems could become grounds for invalidating a patent that has seen non “substantial contribution to the invention.” Naming a human as an inventor where an invention was actually autonomously generated by AI systems could, in certain situations, successfully circumvent the current patent law system, which requires inventors to be humans, given that certain patent offices will not investigate the correct naming of the inventor. Nonetheless, patents issued under such subterfuge are subject to invalidation; the owner will enjoy the same benefits as a legitimate patentee until their patent is invalidated (if caught). But such is a matter of future patent litigation beyond our current topic of analysis.
5. Proposed solutions
The point of friction between the autonomous generation of an invention by AI systems and patent law is that patent law grants a right that is considered a shield. Subsequently, there must be “someone” to wield the shield by enforcing those rights, which machines like AI currently lack. Furthermore, granting patent rights has the fundamental purpose of incentivizing innovation, such could be impaired if owners of AI systems that can autonomously generate inventions acquire many patents and subsequently create a barrier of entry to the market.
Differentiated Status. Firstly, I proposed using the scheme of the spectrum of human intervention to grant different inventorship status (a)human-invention, (b) AI-aided inventions, and (c) AI autonomously generating inventions. Determining the degree of human guidance is necessary regarding how much explanation can be provided within the specification and disclosure of the patent application. Human-invention in which the human inventor uses technologies outside of the AI realm and AI-aided inventions in which the programmer provides explicit tasks to be performed should fall both under the current patent law legal framework because their “nexus” to the human inventor is strong enough as to fulfill the legal standard of human as an inventor. Autonomous generated inventions in which the AI system expands to perform non explicitly programmed tasks should fall within a differentiated inventorship status.
Shorter Period. Subsequently, based on the differentiated inventorship status, patent law should be reformulated in terms of the years of afforded protection. Currently, human-inventions that comply with the patent requirement are granted 20 years of protection. Nevertheless, utility models usually have protection that varies from 6 to 10 years. Even though utility models are not available in certain jurisdictions, utility patens could be feasible to protect inventions autonomously generated by AI systems. The shorter period of protection will be adequate to counterbalance the non-explicit programming nature of certain AI systems. In this regard, those inventions are not able to provide a sufficient disclosure “in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art.” And because of the need to differentiate inventions autonomously generated by AI systems from human-invention to not halter innovations outside the realm of self-improving AI systems.
Similarly, economically relevant factors like the cost of R&D and return investment could be best served in a patent framework with a progressive maintenance fees structure.  If patent costs progress over time, because maintenance fees increase at the end of the patent lifespan, only highly valuable patents (from the patent holder’s perspective) would be kept. Thus, it would address the situation in which autonomous, but not highly innovative, inventions by AI are hoarded by a handful of AI system owners and instead promote innovation by unlocking the invention that may not be valuable for the inventor but can be put into use for a third party to build upon it.
Enablement requirement. Because nowadays turnover times are short economic actors who bring AI to the market take advantage of being first movers instead of relying on patents. Thus, it is crucial to shorten the period of protection while contemporaneously adopting more flexible standards for disclosure of inventions regarding autonomously generated by AI. In terms of sufficient disclosure for granting a 20-year patent, complete enablement a specification needs to include the algorithms used by the programmer and explain the data manipulation process. On the other hand, for granting a patent with a shorter period of protection, disclosure needs to be seen through a flexible enablement standard. Thus, patent applications related to stochastic algorithms it is enough a marginal explanation of the data manipulation process. Such a response could even incentivize programmers to develop solutions for the black box problem. In these terms, the more disclosure they can provide (not willing to provide) would grant them more years of protection exclusivity.
6. Which rights?
It is important to clarify what rights are granted when obtaining inventorship status related to AI-aided inventions and to AI autonomously generating inventions. Using the scheme of independent rights in combination with the concept of legal personality could allow pursuing policy goals like encouraging entrepreneurship and contribute to the coherence and stability of the legal system. Independent rights are used in the context of “invention” to convey the idea that something is invented “free from influence guidance or control of another or others.” While legal personality is used to identify an entity to which to assign responsibility, juridical persons are non-human entities granted certain rights and duties by law. In this case, the differentiated inventorship status avoids disincentivizing human-invention and deter falling into the “android fallacy,” the tendency to describe AI systems in human capacity terms. As widely known, the critical question is not whether we can but whether we should. A reason for considering whether AI systems should be recognized a differentiated inventorship status should be focused, not on what they are, but what they can do.
In the case of differentiated inventorship status, the AI autonomously generated inventions would provide the right to the AI system of being recognized as a differentiated status of inventorship, a more flexible standard for disclosure but a shorter period of exclusivity. The duty would be to explicitly acknowledge the non-explicitly programmed nature of the AI system and the autonomous nature of the process behind the final invention.
The programmers and users of the AI system could be granted “parallel moral rights.” Moral rights connect the author to his work. French copyright law provides the author with four principal attributes: the right to disclosure, right to attribution, right to withdrawal and retreat, and right to the integrity for protected works. Contrary to copyright, we cannot pretend to have any prerogative outside the patent because it is hard to contest that a certain innovation was triggered by a creative process and is imprinted with the inventor’s personality. It is especially hard to contest that AI-autonomously generated inventions imprint the programmer’s personality in their invention. However, expressing intelligence can be regarded as expressing personality. Thus, in areas that require enormous expressions of intelligence in the form of expertise and technical skills, like programming self-improving software that autonomously generated invention, the programmer can be said to imprint its personality/intelligence in the “automated model fitting” AI system. This imprinting of personality/intelligence might not be enough to grant the programmer with inventorship status but requires such levels of personality/intelligence as to grant “parallel moral rights” like disclosure, withdrawal and retreat, and integrity.
The field of law, in particular legislation, has always been regarded as not able to keep pace with technological innovation. Such creates uncertainty and halters investment in innovation. Thus, a growing interest in anticipating the future is a shift for the better. Nonetheless, the tendency to describe AI systems in human capacity terms requires questioning not whether we can grant inventorship but whether we should. Because in the current state of the art regarding AI systems, we are far from “strong AI,” the current efforts should focus on two goals. First, understanding how to adapt existing legal frameworks to the emerging AI systems and the outputs they are generating. It follows that we should grant AI systems inventorship that is clearly different and focuses not on what they are but what they can do. Secondly, thinking about what modifications to the current legal framework are needed to best address a future that is arriving. Such will require evaluating solutions inspired by other fields of law and expanding concepts of different areas of IP. For such, I advocate for the combination of concepts of legal personality and independent creation to grant inventorship to AI systems and extending concepts of moral rights to the patent law system to allow programmers to lean on their personality/intelligence contribution to the AI systems and retain rights over AI-autonomously generate inventions.
 Martin Ford, Architects of Intelligence: The Truth About AI from People Building It. 6 (Dominic Shakeshaft ed., Packt Publ’g Ltd. 2018).
 World Econ. Forum & Ctr. for the Fourth Indus. Revolution.
 Daria Kim, ‘AI-Generated Inventions’: Time to Get the Record Straight?, 69 443, 446-47 (2020) https://academic.oup.com/grurint/article/69/5/443/5854752.
 High-Level Expert Group on Artificial Intelligence, A Definition of AI: Main Capabilities and Disciplines, Eur. Commission 6 (2019), .
 Josef Bajada, Artificial Intelligence Demystified, Towards Data Sci. (2019), https://towardsdatascience.com/https-medium-com-josef-bajada-demystifying-artificial-intelligence-6f5f7a8dd1b0.
 World Econ. Forum & Ctr. for the Fourth Indus. Revolution.
 Kim, supra note 3, at 444-446.
 Request for Comments on Patenting Artificial Intelligence Inventions, 84 Fed. Reg. 44889 (2019) https://www.federalregister.gov/documents/2019/08/27/2019-18443/request-for-comments-on-patenting-artificial-intelligence-inventions.
 Id. at 51522.
 Dilip Kumar Pratihar, Intelligent Autonomous Systems: Foundations and Applications (Lakhmi C. Jain ed., 1st ed. Springer Verlag 2010).
 Yampolskiy R.V. (2015) Analysis of Types of Self-Improving Software. In: Bieger J., Goertzel B., Potapov A. (eds) Artificial General Intelligence. AGI 2015. Lecture Notes in Computer Science, vol 9205. Springer, Cham. https://doi.org/10.1007/978-3-319-21365-1_39.
 Hoos H.H., Stützle T. (2015) Stochastic Local Search Algorithms: An Overview. In: Kacprzyk J., Pedrycz W. (eds) Springer Handbook of Computational Intelligence. Springer Handbooks. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-662-43505-2_54.
 U.S. Patent Application No. 5,852,815A (issued Dec. 22, 1998).
 Jonathon Keats, John Koza Has Built an Invention Machine, 2006 , Apr. 19, 2006 at (2006) available at https://www.popsci.com/scitech/article/2006-04/john-koza-has-built-invention-machine/.
 U.S. Patent Application No. 6,847,851 B1 (issued Jan. 25, 2005).
 The Artificial Inventor Project, Patent Applications, (2019) https://artificialinventor.com/patent-applications/.
 Supra note 15.
 In Re Application No.: 16/524, 350, available at https://www.uspto.gov/sites/default/files/documents/16524350_22apr2020.pdf?utm_campaign=subscriptioncenter&utm_content=&utm_medium=email&utm_name=&utm_source=govdelivery&utm_term=.
 Intellectual Prop. Office, available at .
 EPO Publishes Grounds for Its Decision to Refuse Two Patent Applications Naming a Machine As Inventor, Jan. 28, 2020 available at https://www.epo.org/news-events/news/2020/20200128.html.
 Supra, note 23, The tribunal observed that Thaler could not have acquired ownership from DABUS ‘as the inventor cannot itself hold property’ (para 23).
 Thaler v. Commissioner of Patents  FCA 879. Available at https://www.judgments.fedcourt.gov.au/judgments/Judgments/fca/single/2021/2021fca0879.
 Notice of issuance was published in July 2021 Patent Journal of South Africa’s Intellectual Property. Available at https://www.ipwatchdog.com/wp-content/uploads/2021/07/AP7471ZA00-Notice-of-Acceptance-1.pdf.
 Garrett Corp. v. United States, 422 F.2d 874, 881 (Ct. Cl. 1970).
 Jon Kleinberg et al., Discrimination in the Age of Algorithms, 2019 Nat’l BUREAU of Econ. Res. 19 (2019) available at https://www.nber.org/system/files/working_papers/w25548/w25548.pdf.
 Kim, supra note 3, at 445.
 Robert Plotkin, The Genie in the Machine: How Computer-automated Inventing Is Revolutionizing Law and Business 3 (Stanford Univ. Press 2009).
 Josef Drexl et al., Comments of the Max Planck Institute for Innovation and Competition of 11 February 2020 on the Draft Issues Paper of the World Intellectual Property Organization on Intellectual Property Policy and Artificial Intelligence, 3 available at https://pure.mpg.de/rest/items/item_3193085_1/component/file_3193086/content.
 Kleinberg, supra note 28.
 Supra, note 30.
 Rich Sutton, The Bitter Lesson, Mar. 13, 2019 at available at http://www.incompleteideas.net/IncIdeas/BitterLesson.html.
 Anne Lauber-Rönsberg, Sven Hetmank, The concept of authorship and inventorship under pressure: Does artificial intelligence shift paradigms?, Journal of Intellectual Property Law & Practice, Volume 14, Issue 7, July 2019, Pages 570–579, https://doi.org/10.1093/jiplp/jpz061.
 High-Level Expert Group on Artificial Intelligence, Ethics Guidelines for Trustworthy AI (European Comm’n 2019), https://ec.europa.eu/digital-single-market/en/news/ethics-guidelines-trustworthy-ai.
 Daria Kim, ‘AI-Generated Inventions’: Time to Get the Record Straight?, 69 GRUR Int’lIssue 5 443, 448 (2020), available at https://academic.oup.com/grurint/article/69/5/443/5854752.
 Scott Yackey, Why Listing All Inventors Is Crucial to Protecting an Invention, Dall. Bus. J., Mar. 5, 2019 at (“Errors in identifying the inventors can render a patent unenforceable, invalid, or alter ownership”) available at https://www.bizjournals.com/dallas/news/2019/03/05/why-listing-all-inventors-is-crucial-to-protecting.html.
 Abbott, supra note 17, at 1081.
 Rule 19(2) and Article 60(3) EPC (“[f]or the purposes of proceedings before the [EPO], the applicant shall be deemed to be entitled to exercise the right to the European patent”). Note that, nonetheless, Article 60(1) EPC states that, identifying the inventor of a given invention is very important since “[t]he right to a European patent (…) belong[s] to the inventor or his successor in title”
 See Kennedy v. Hazelton, 128 U.S. 667, 672 (1888) (“A patent which is not supported by the oath of the inventor, but applied for by one who is not the inventor, is unauthorized by law, and void, and . . . confers no right as against the public.”).
 James Bessen, A Generation of Software Patents, 18 Bos. U. J. of Sci. & Tech. L. 241, 248 (2012) (“When firms acquire large numbers of patents, they can restrict entry into an industry and they can use these patents to extract rents from other fines beyond the rents needed to encourage innovation.”) available at https://www.hbs.edu/faculty/Shared%20Documents/conferences/2014-strategy-research/A%20Generation%20of%20Software%20Patents.pdf.
 McLaughlin, Computer-Generated Inventions, at 20.
 Kim, supra note 3, at 450-451.
 McLaughlin, supra note 43, at 23.
 Kim, supra note 3, at 451.
 Agreement on Trade-Related Aspects of Intellectual Property Rights, art. 33, Apr. 15, 1994 (“The term of protection available [for patents] shall not end before the expiration of a period of twenty years counted from the filing date.”); The European Patent Convention, art. 63(1), Mar. 20, 1883 (“The term of the European patent shall be 20 years from the date of filing of the application.”); 35 U.S.C. § 154(a)(2) (“Subject to the payment of fees under this title, such grant shall be for a term beginning on the date on which the patent issues and ending 20 years”)
 World Intellectual Prop. Org., Utility Models, available at https://www.wipo.int/patents/en/topics/utility_models.html.
 IP Helpdesk, What Can Be Protected By a Utility Model?, Eur. Union, Dec. 15, 2020 at (“Because the novelty and inventive step requirements are less stringent than those required for patents, utility models can be useful for incremental inventions where only a small change has been made and which might not meet the inventive step requirements for a patent.”) available at .
 The European Patent Convention, art. 83, Mar. 20, 1883 (“Disclosure of the invention”) available at https://www.epo.org/law-practice/legal-texts/html/epc/2016/e/ar83.html; 35 U.S.C. 112(a) (“[Patent] specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.”)
 Benjamin N. Roin, The Case for Tailoring Patent Awards Based on Time-to-Market, 2014 UCLA L. Rev. (2014) available at https://www.uclalawreview.org/pdf/61-3-3.pdf; Everis Spain S.L.U., Patent Costs and Impact on Innovation. International Comparison and Analysis of the Impact on the Exploitation of R&D Results By SMEs, Universities and Public Research Organisations. , 2014 Eur. Commission (2014) (Arguing that patent grants an exclusive right to exploit an invention in a certain market, patent costs and returns need to be measured in relative terms to the size of the market where the protection is sought.) available at https://ec.europa.eu/jrc/communities/sites/jrccties/files/patent_costs_and_impact_on_innovation.pdf.
 Everis Spain S.L.U., Patent Costs and Impact on Innovation. International Comparison and Analysis of the Impact on the Exploitation of R&D Results By SMEs, Universities and Public Research Organisations., 2014 51 (2014) (economic value of patents…best indication of the economic value of such assets are the accomplishment of a deal for a certain price or the exploitation of a new product (or the improvement of current ones) being sure of the contribution of the corresponding patents to these products) available at https://ec.europa.eu/jrc/communities/sites/jrccties/files/patent_costs_and_impact_on_innovation.pdf.
 Id. at 8, 22, and 39.
 Bessen, supra note 42.
 Everis Spain S.L.U., Patent Costs and Impact on Innovation. International Comparison and Analysis of the Impact on the Exploitation of R&D Results By SMEs, Universities and Public Research Organisations., 2014 51-53 (2014) (“Patents are mostly filed by strategic or internal motivations (new product development in the case of enterprises) rather than for licensing purposes…being part of a patent family is more relevant than being granted [a patent]…[because] to be part of a set of patents allows the patent to add value to its “relatives” and, at the same time, its perceived value is also increased”) available at https://ec.europa.eu/jrc/communities/sites/jrccties/files/patent_costs_and_impact_on_innovation.pdf.
 Shlomit Yanisky Ravid and Xiaoqiong (Jackie) Liu, When Artificial Intelligence Systems Produce Inventions: An Alternative Model For Patent Law At The 3a Era, 39 Cardozo L. Rev. 2215, 2255 (2017) available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2931828.
 Kim, supra note 3, at 451.
 Simon Chesterman, Artificial Intelligence and The Limits Of Legal Personality, 69 819, 823 (2020) available at doi:10.1017/S0020589320000366.
 Intellectual Property Expert Group, Independent Invention, (2009) available at https://www.ipeg.com/independent-invention/.
 Gabriel Hallevy, Liability for Crimes Involving Artificial Intelligence Systems (Springer Int’l Publ’g Switz. 2014) (arguing that it is AI systems’ rationality that provides the basis for personhood).
 Chesterman, supra note 58. (Noting that the content of legal personality is very variable, it can be just rights without obligations or just come with obligations.)
 Paris Convention for the Protection of Industrial Property, art. 4ter, as amended on September 28, 1979. (“The inventor shall have the right to be mentioned as such in the patent.”); Jeremy Philips, Right to Be Known As Inventor. Is Xu Rong to Sue? (2012) (“But there is no right to object to false inventorship and no right to be named as inventor outside the four corners of the granted patent; nor is there any right to be acknowledged as the founding father of a new area of science of technology.”), https://ipkitten.blogspot.com/2012/12/is-xu-rong-to-sue.html.
 Bertrand Sautier, Moral Rights in Patent Law: Oxymoron Isn’t It?, (2013) available at https://ipkitten.blogspot.com/2013/09/moral-rights-in-patent-law-oxymoron.html.
 Code de la propriété intellectuelle, Droits moraux (Articles L121-1 to L121-9) available at https://www.legifrance.gouv.fr/codes/id/LEGIARTI000006278891/1992-07-03/.
 Sautier, supra note 65; Nicolas Bronzo, Le droit moral de l’inventeur, Propriété Industrielle, juin 2013, p. 9 et s.
 Sautier, supra note 65.