Материал: Том 1 №1
Digital Law Journal.Vol.1, No.1, 2020,p.24–38
ElinaL.Sidorenko,PierrevonArx/TransformationofLawintheContextofDigitalization
Results
The paper analyses the needs and requirements for adapting the rule of law, in order to address the legal challenges arising from the development and the use of digital technologies and their products. It considers the categories of legal personality and protection of digital technologies, as well as the foundations of their legal regulations. The paper defines ways to protect intellectual property rights in the digital sphere, and considers the development of legislation in terms of establishingtheresponsibilityofbothdevelopersandusersofdigitalprograms.
In addition, the article reveals the author’s model for considering the legal status of digital technologies through the prism of general issues of status regulation and establishing responsibility and applied aspects of the application of individual technologies (Artificial Intelligence, Internet of Things,Blockchain,bigdata,etc.).
The paper notes the trend in the digital sphere of law being marginalized, and emphasizes the limited adaptive capabilities of traditional legislation in the regulation of digital technologies. The authors propose the development of fundamentally new legal constructions for digital technologies regarding their legal personalities and ability to be protected; this would allow the goal of the progressive development of the digital economy and the technical capabilities of individual technologiestobetakenintoaccount.
Discussion
Modernscientificliteraturedoesnotaddressthelegalstatusofdigitaltechnologies,ortheuseof itsproducts,inaholisticmanner.Asillustratedbelow,expertsfocustheirmainattentiononspecific technicalissuesandpracticalsolutionsconcerningparticulartopics.
Forexample,theissuesoflegalregulationinthesphereofArtificialIntelligencepredominantlysit withintheframeworkofthetechnologyreliabilityevaluation(Yu&Alì,2019);itspossibleapplication incertainareasofactivity,inparticularinjurisprudence(Mowbrayetal.,2019);orincitingchangesin modern tort and contract legislation (Hacker et al., 2020). Specialists mainly address issue of recognizingAIasanobjectofcivilrights,andanobjectorasubjectofintellectualactivity.
The legal regulation of drones and other breakthrough technologies’ application is addressed with regard to specific legislation reform (aviation (Bassi, 2020), transportation (Bassi, 2020), health care(Konertetal.,2019),information(Marquès,2019)andothers.Inscientificliterature,greatimportance is attached to the regulation of smart contracts in terms of securities exchange (Lee & Joseph, 2019) declaration of the parties’ intent (Gomes & Silvana, 2018), and de jure formalization of smart contracts(Liu&Huang,2019).
Cryptoassetsmanagementisalsoaddressed.Themajorityofworksaredevotedtobasicfinancial lawprinciples(Giudicietal.,2020),theregulationofthedigitalpaymentinstrumentsmarket(Huang etal.,2020),taxation(Sixt&Himmer,2019)andnorm-settingprinciplesinthefieldofdigitalfinance (Edwards,Hanley,Litan&Weil,2019).
The issue of the status of big data legal is equally urgent; examples of these issues include its use in legal activities (Goanta, 2017;Custers&Leeuw,2017),legalcontrol(Lei,2019),andmodelsand assessmentoflegalrisksrelatedtobigdatause(Low&Mik,2019).
30 ARTICLES
Цифровоеправо.Том1, №1, 2020, с.24–38
Э. Л.Сидоренко,П.фонАркс/Трансформацияправавконтекстецифровизации
We shall not attempt a detailed analysis of all legal challenges related to digitalization, but rather outline two clusters of issues that need to be addressed as a priority, both at national and international levels. The first cluster shall focus on the general aspects of transformation of digital law: adaptivecapacitiesofmodern,civil,financiallegislationindigitaltransformation;thelegalnatureof digitaltechnologiesasanobjectofcivilrights;andanyspecificitiesofcontractualrelationsindigital economy. The second cluster shall focus on the implementation of particular technologies, such as the Internet of Things, AI, big data, machine learning, drones, robots, and similar areas, which require detailedscientificanalysisandrestrictiveapplication.
While assessing the prospects for the development of law in the context of digitalization, scientistsareconsideringtwopossiblescenarios:thefirstoneforeseestherestructurationoftraditional legal models in order to accommodate new digital issues; the second one foresees fundamental changesinthecurrentlegislationaimingatreplacingtraditionallegalframeworkwithmoreabstract anduniversalmodels.
The conflict between the two scenarios is most visible in the analysis of protectability and legal identityofdigitaltechnologies.Protectabilityofdigitaltechnologiesmeanstheirabilitytoactasobjects of civil and intellectual rights. In this regard, it is important to consider whether emancipation ispermissiblebetweentherealright/rightinremandcopyrightfromtraditionallegalinstitutions.
Emancipation opponents challenge the fact that Artificial Intelligence and machine learning can become the foundation for the law, or they can replace fundamental principles of the law, as these principleshavedevelopedoverthecenturiesthroughso-called‘consolidatinglearning’inAI,whilst humanbehaviourisinherentlytooirrationalandinconsistent(Fernández-Villaverde,2020).
Proponents, by contrast, see emancipation as a natural conflict of public and private interests in the digital world (Entin, 2017) and recognize the possibility for digital technologies to obtain their legalidentityiftheyacquirefunctionalautonomy,value,andeconomicutility(Kharitonova,2019).
In the current Russian legislation, objects of civil rights, along with property rights, include digital rights. They constitute a type of property rights and are mentioned along with intellectual rights, which provides some researchers with a reason to deny the possibility of attributing digital objects to copyright protection. They are defined as the right of obligation or other rights; their exercise, disposal, or restriction is possible only within the information system (Article 141.1, the Civil Code of the Russian Federation) and therefore they have a very limited meaning or application. A whole clusterofissuesrelatedtoexercisingtheserightsoutsidethedigitalenvironment—namely,results of the rights implementation (big data consolidation and analysis, machine learning results, etc.) — remainsoutsidetheregulatorymechanisms.
While Russian law provides for the regulation of digital rights as part of information law, in English law the protection of digital rights is related to the protection of property. This approach is in line with the spirit of the following statement: Digital rights are monetary in nature and therefore should be protected as things (Rahmatian, 2013).Fromthispointofview,digitaltechnologiesacquire thestatusofobjectsofcivillaw,aswellasproductsmanufacturedusingthesetechnologies.
Thus,theAnglo-Saxonlawmakesiteasytoanswerthequestionsaboutwhoowns(a)anewcode of a self-learning program, which embodies ‘experience’ in executing commands; (b) the rights to products made by such programs and robots; and (c) who bears property or other liability for the
СТАТЬИ 31
Digital Law Journal.Vol.1, No.1, 2020,p.24–38
ElinaL.Sidorenko,PierrevonArx/TransformationofLawintheContextofDigitalization
negativeeffectsofdigitaltechnologiesortheirproducts.However,othercountries’experiencesare not sufficient for the sustained and successful development of the digital economy. It is important todevelopaone-size-fits-allapproachforcivilmatterswithregardtodigitaltechnologies,products, andrights.
We believe that this approach should be based on the following methodological provisions:
1) Recognition of digital product as an object of civil rights shall not be based on its material essence or economic value. As a rule, digital objects have a comprehensive legal nature and can be considered both from the intellectual and the proprietary standpoint;
2) Digital technologies’ assessment shall be based on a legal model, allowing for compliance with the legal nature of the relationship, and which is capable of balancing private and public interests in digital circulation;
3) When developing a universal approach to digital technologies regulation as a subject of civil rights, objective emancipation of digital law from traditional legal institutions shall be taken into account. In other words, the adaptive capacity of traditional law in relation to digital technologies isverylimitedduetothemultifunctionalnatureofobjects,technicalsaturation,uncertaintyintask management,andpossiblerisksrelatedtotheirintegrationintocivilcirculation.
With regard to the development of intellectual rights in the digital sphere, it is important to note thatoneofthemostpressingissuesisthechoicebetweengrantingdigitalbenefitstoindividualsor recognizing them as public domain. However, it is obvious that it is not possible to take a decision as long as the legal nature of the rights of digital products is not determined; this is particularly the caseconcerningAI.
Challenges emerge from the objective characteristics of digital products: technology results being highly repeatable, the low human creative contribution, the automation of some processes, the impossibility to distinguish between creative and noncreative components, the complexity of distinguishingbetweentheauthor’srightsandthoseofthecompiler,andsoon.Suchchallengesarealso encountered while referring to digital platforms and platform solutions, databases, data processing algorithmscreatedduringmachinelearning,etc.
The digital economy poses a question for the law on how to protect intellectual activity products and fully respect the interests of creators, users, and investors. Three possible models related to intellectual products’ protection are considered by scientists and in practice:
1. Consider digital technology products through the prism of copyright. This approach is based on European Union Directive 96/9/EC (EC, 1996) and is valid for cases where digital objects meet the criterion of originality for data selection and processing. However, this approach negates the difference between the author and the technology owner. In addition, there may be difficulties in distinguishing authorship when, for example, one author creates their product based on another author’sdigitalsolution(digitalplatform,blockchainregistry,cloudtechnology),orwhensomedata istransferredfromthecreator’sdatabasesystemtoanothersystemwithouttheirconsent.Moreover, theproposedsolutionisinconflictwithoneofthebasiccopyrightprinciples—protectingaproduct that has an original, unchangeable, and one-off form. Digital solutions typically have multiple pre-
32 ARTICLES
Цифровоеправо.Том1, №1, 2020, с.24–38
Э. Л.Сидоренко,П.фонАркс/Трансформацияправавконтекстецифровизации
sentation formats, coding methods, and techniques, which precludes considering them as indivisible originalproducts.
2. Digital technologies regulation in neighbouring rights’ format. This approach is common for
Russian law, where the activity of database or other digital products creators can be determined as organizational and technical (Maggon, 2006). In this case, database architecture creators and developersacquireanexclusiverighttotheirintellectualproductsingeneral.However,theboundariesof neighbouringrightsareblurred,andtheriskofdigitalproductsusers’rightsbeingviolatedincreases.
3. Establishing an independent institute of intellectual rights in digital objects, combining both property and non-property rights, as well as reflecting the technological specificity of protected objects (Lauts, 2019). The development of a fundamentally new mechanism will allow a correct model formanagingdigitalproductsusetobedesigned(productscreatedbyrobots,self-learningprogram codes,etc.).
Thenextimportantaspectistoprovidealegalcapacitytodigitaltechnologies.Thisproposalhas already exceeded its theoretical framework. There are proposals to recognize robots and robotics products as subjects of rights of obligation and to consider them as agents in concluding contracts with third parties acting on behalf of their owner and in their own name (Neznamov & Naumov, 2008).Arizonalawgivesdeliveryrobotssamerightsaspedestrians,buttheymustabidebythesame rules: they cannot run into somebody and must give way to other pedestrians. The state of Utah is currentlyconsideringabillwiththesamewording.
The traditional law identifies legal identity as a combination of legal capacity, active capacity, and delictual dispositive capacity. Currently, three types of legal capacities are recognized: physical persons, corporations, and public legal entities. At the same time, this list has gradually expanded, as necessary, to include new circulation parties and to adjust the theoretical foundations of legal identity. In particular, granting legal identity to corporations requires changed approaches to understanding the will, interest, and motive of the subject. Moreover, it does not seem appropriate to denydigitaltechnologies’legalidentityduetothelackofawillcomponent(Ponkin&Redkina,2018).
It is much more important to evaluate technologies through the prism of corporations’ legal iden- tity:namely,fromthestandpointofautonomyanddecision-making.Learningabilityandtheability toindependentlychangetheactionalgorithmcansupplementthesecharacteristics.
However,itshouldbenotedthatmoderndigitaltechnologieshavedifferentdegreesofautonomy.
Can the same legal identity rules apply to them, or should they be differentiated? In modern law, thereisnosolutiontothisissueyet.IfthelegalidentityofAIandotherdigitaltechnologiesisrecognized, it will be important to address such issues as public liability insurance, criteria for determining thepossibleriskrelatedtoroboticactivities,settingouttherulesfordigitaltortandtheprocedure forrecording,andregisteringnewlegalentities.
In general, the very idea of the legal identity of digital technologies is reasonable; however, it oftenequatedwiththelegalidentityofpersons.Accordingtoexperts,thelegalidentityofrobotsmay be recognized similarly to the recognized legal identity of international organizations, for example theUN(Chung&Zink,2017).
СТАТЬИ 33
Digital Law Journal.Vol.1, No.1, 2020,p.24–38
ElinaL.Sidorenko,PierrevonArx/TransformationofLawintheContextofDigitalization
It is also important to consider the autonomy of digital objects. Human dependence on technology will determine its delictual dispositive capacity. It is important to mention a number of applied aspects with regard to law transformation in the digital environment. In particular, AI application issues need to be resolved. It is difficult to identify persons who are responsible for digital device errors. Experts widely discuss the liability of IBM’s Artificial Intelligence (AI) system “Watson for Oncology” in South Korea, and its legal identity, and argue that liability should be placed on Watson’s creatorsandtherelevantmedicalpersonnel.
Establishment of legal guarantees for AI and robots security is widely discussed alongside activity regulation issues. In particular, Chessman raises the issue of applying animal handling rules to robots—uptotheestablishmentofresponsibilityfortheabuseofrobotsorAI—ascontemplation of this abuse can cause mental harm to humans (Chessman, 2018).The active introduction of AI and machine learning makes the process of determining any possible limits for the use of technology in legalactivitiesachallenge.Someauthorsdenythispossibility,andstressthatlegaldecision-making cannot be automated as it contains too many value judgments; norms are contextual, and reliant on intuition as well as fairness in any of its forms (Wachter et al., 2020). This approach seems to be extremely categorical, especially in light of the long-standing use of digital technologies by lawyers.
The issue is how this participation shall be formalized, and how to determine the machine’s liability forthefinaldecision.RyanCatterwell’sfaircommentwasthatmachinelearningandAIcouldbeused forautomatedcontractinterpretation.
However, if some provisions may well be interpreted by a machine, others may not. There are two main constraints: 1) some provisions are relative, intuitive, and rather a “question of perception”; 2) some provisions can be interpreted only in the light of the views of the parties and their circumstances. Therefore, machine learning can only help lawyers, not replace them in contract interpretation (Catterwell, 2020). The separation of human and machine power in decision-making is therefore justified. AI can assume the function of data processing and analysis, and a person can makedecisionsbasedontheircriticalanalysisofinformationpreparedbyAI.
Many issues also arise while using the Internet of Things. Lawyers emphasize that the personal data which users of “smart” things provide is insufficiently protected; devices often collect and exchange information without the owner being aware of it, or can exchange information across borders, amongst other violations. (Mohamed & Zulhuda, 2015). The problem is that the current legislation, andinparticulartheEUGeneralDataProtectionRegulations(GDPR),doesnottakeintoaccountthe use of the Internet of Things and smart homes; literal application of regulations may impose an unreasonably high liability on device designers in some cases, or may create cyber security risks in othercases(Chenetal.,2019).
Theissueofpersonaldataprotectionalsoarisesinconnectionwiththeuseofcloudtechnologies. Even the current strict General Data Protection Regulation (GDPR) cannot be applied in practice for personal data processing in cloud technologies, in particular due to its inability in identifying data controllers or providers; moreover, the fact that the programme has collected data may remain secret(FoschVillaronga&Millard,2018).Theexchangeofdatainthedatacloudisnolessproblematic. Currently, much data — including the data needed for criminal investigation by Country-1 — may be under the control of Country-2. However, international mutual legal assistance treaties do not
34 ARTICLES