Is it possible to grant legal personality to artificial intelligence software systems?

doi:10.1016/j.clsr.2017.03.022

Computer Law & Security Review

Volume 33, Issue 5, October 2017, Pages 685-699

https://doi.org/10.1016/j.clsr.2017.03.022 Get rights and content

Abstract

The purpose of this paper is to determine whether Systems of Artificial Intelligence (SAI) can be deemed subjects of law. This aim is formulated according to the technical capabilities integrated in SAI and the SAI's ability to interact independently with other legal subjects. SAI features, such as direct connection with intellectual skills, the ability to understand, learn and make autonomous decisions may cause situations where autonomous systems based on AI will make decisions which will be in the best interests of individuals, even though conflicting with the user's own will.

To consider the possibility of SAI being recognized as possessing legal personality, we analyse the concept and features of SAI and define its operating principles. We give hypothetical examples to demonstrate the necessity of SAIs being recognized as such. The paper undertakes legal personality analysis of SAI performed: (i) using the philosophical and legal concepts of a subject (person); (ii) discussing artificial (unnatural subjects of law) as an alternative to the recognition of legal personality of SAI; (iii) using elements of legal personality set for natural and legal persons.

The analysis leads to the conclusion that the scope of SAI rights and obligations will not necessarily be the same as the scope of rights and obligations of other subjects of law. Thus, SAI could only have rights and obligations that are strictly defined by legislators. This conclusion suggests that the result of this paper may be its use in further research defining the scope of SAI rights and obligations.

Introduction

“Can a machine think?” The question attracting more and more attention of scientists and practitioners was raised back in 1950 by Alan Turing¹ who thereby set a direction for the discourse on Artificial Intelligence (AI). AI is rather new discipline, which has no single definition yet. The concept of AI was first mentioned in 1956²; the system is often defined as artificially developed intelligence related to rapidly developing technologies, which enable computers to operate intelligently, i.e. in a human-like manner.³ Systems of Artificial Intelligence (SAI) are different from other regular computer algorithms (programs) due to their uniqueness, since they are able to learn independently, gather experience and come up with different solutions based on the analysis of various situations independently of the will of their developer (programmer), i.e. SAI are able to operate autonomously rather than automatically.

Today, there is a great variety of information technologies based on the operating principle of SAI, for example, Google Self-Driving Cars, autopilots controlling airplanes, digital assistants such as Siri, Cortana and Google Now, robot nurses, mind-controlled Google smart glasses, etc. These and other technologies are known worldwide and their capacity as well as use is rapidly growing. As the use of technologies based on AI becomes more and more widespread, the number of associated incidents grows as well. For example: (i) in 1.7 million miles of travel, Google Self-Driving Car had 11 accidents resulting in damage⁴; (ii) speech recognition software could become a contributory factor to car accidents⁵; (iii) robot nurses reminding patients to take their medicines fail to ensure that the medicines are actually taken, which may lead to the patient's death.⁶

These specific examples show that SAI are not mere science fiction.⁷ Information technology innovation based on SAI and the above examples allow stating that SAI are not mere objects, operation of which is influenced by others. SAI act like entities.⁸ Regardless of the exceptional operating principle of such systems, none of the legal systems has recognized SAI subjects of law so far. However, is such legal status of SAI only a temporary attribute, which should change in time? Is it possible to grant legal personality to a System of Artificial Intelligence? Even though the problem at hand has already been addressed before,⁹ it has attracted greater (proper) attention only in the recent years.

It should be noted that SAI that are capable of learning and making decisions independently can make the lives of people easier, but failure to manage such technology can lead to major existential threats.¹⁰ Bill Gates claims that after a few decades, SAI and the level of their intelligence will lead to major concerns; therefore, it is necessary properly to prepare ourselves.¹¹ As the role of SAI in our daily lives becomes more and more important, we encounter various challenges: moral, ethical issues and problems. Legal regulation and legal system itself are not an exception. Discussions on the status of SAI that are able to make more and more complicated decisions independently in the legal systems of countries become increasingly frequent and extensive at the academic level,¹² in the political arena of various countries¹³ as well as in the context of its shaping, and in the society.

The ability of SAI to learn from their own personal experience leads to independent conclusions and autonomous decision-making, i.e. what can lead them to their legal personality. Due to their ability to make decisions independently, technologies based on such systems like Machine Learning, Expert Systems or Neural Networks can no longer be treated as objects. Therefore, the aim of the paper is to determine whether SAI can be deemed subjects of law. The object of the paper is legal personality of SAI and the methods of research are information collection, systematizing, generalizing, valuation, comparison, analysis of scientific literature, synthesis and deduction.

The paper consists of five sections. Section 1 presents the concept of SAI and analyses the concept and main features of such system, and defines its operating principle. Section 2 answers the question whether it is necessary to recognize SAI as subjects of law. Section 3 analyses legal personality of SAI. The analysis is performed using the philosophical and legal concept of a subject (person). Section 4 discusses artificial (unnatural) subject of law as an alternative to the recognition of legal personality of SAI. Section 5 analyses the legal personality of AI using elements of legal personality set for natural and legal persons.

Section snippets

Revealing the concept of a system of artificial intelligence through the use of definitions of artificial intelligence

AI is a new discipline, which covers an extensive field. Therefore, it has no single definition. Development and research of SAI began with the emergence of the first personal computers.¹⁴ John McCarthy first mentioned the concept in 1956.¹⁵ Even though the use of SAI is becoming more and more extensive in the everyday lives of many people, it is still rather difficult to come up

Is it necessary to recognize SAI as subjects of law?

The issue of possibly granting legal personality to various objects, including SAIs, is becoming increasingly relevant in the public domain. For example, on 20 April 2015, representatives of the Nonhuman Rights Project organization announced that, for the first time in history, the Manhattan Supreme Court had recognized two chimpanzees (Hercules and Leo) as legal persons.²⁵

Legal subjectivity of SAI according to philosophical and legal concepts of the subject

Once we arrive at the conclusion that recognizing SAI as subjects of law would facilitate involvement of these technologies in the civil and legal circulation, another important question arises: could a System of Artificial Intelligence, as such, be recognized as a subject/field of law in legal terms? One of the definitions of subjectivity given by the Oxford English Dictionary is consciousness of one's perceived states. In philosophical terms, subjectivity is defined as an individual's

Artificial subjects of law in legal systems

Existing civil law and common law traditions recognize two types of subjects of law: natural persons and legal entities.⁸²

Legal personality of artificial intelligence based on the elements of legal personality attributed to natural persons and legal entities

The ability of an entity to acquire rights is one of the characteristics defining the subject of law.¹⁰⁷ However, making the legal status conditional on the mere ability of the entity to have rights would excessively expand or narrow down the list of subjects of law, as all living entities capable of feeling would either be treated as subjects of law or, on the contrary as inanimate entities,

Conclusions

Artificial Intelligence is a rather new discipline covering a wide area. Therefore, there is no single definition of the concept yet. In their papers, various authors define the concept of Artificial Intelligence as artificially created intelligence, for example, a software system able to imitate human thinking processes with the help of computers or other devices. Stuart Russel and Peter Norvig distinguish two main directions for the concept of Artificial Intelligence: (i) related to thinking

Paulius Čerka, Attorney in Čerka and The Partners advocates bureau, Associate Professor; Dr., Vytautas Magnus University Faculty of Law, Department of Private Law, Kaunas, Lithuania.

References (0)

Cited by (72)

Artificial intelligence in the work process. A reflection on the proposed European Union regulations on artificial intelligence from an occupational health and safety perspective
2023, Computer Law and Security Review
Artificial intelligence (AI) finds increasingly growing applications in the working environment. Its importance has been recognised by the European Parliament and the European Commission, as reflected in the legislation prepared at the European Union level. As the use of AI creates new risks hitherto unknown from an Occupational Health and Safety (OHS) perspective, the question is whether the proposed EU regulations address these risks. The starting point for further consideration should be an analysis of the proposed changes to EU law in the context of the general principles of labour law. In addition to proposals to amend EU law on artificial intelligence, this article examines current occupational safety and health legislation. Issues related to occupational safety and health monitoring of employers using artificial intelligence were also the subject of the study. The social sciences' perception of human labour is not insignificant in assessing the new relationship at the employer-AI-employee level. The proposed model for regulating AI by the EU legislator is insufficient. First and foremost, there is no clear indication of the employer's obligations towards employees concerning occupational health and safety. Certainly, the essence of the EU law should be to establish the role of the employer in the process and of the Labour Protection Authorities, on the assumption that AI is only a working tool and not a subject of the law. Authorities should work together with employers to achieve the regulation objectives. Consideration should be given to introducing a responsive method of regulation in EU law, whereby the employer's application of employee health protection standards would be reviewed by the Authority to ensure that the objectives of the regulation are met. Depending on achieving a specific outcome or failing to meet the set targets, the Authority could intervene by applying appropriate punitive, incentive or reward mechanisms to the employer. Such a system of action, supported by providing adequate information to employers about the new risks associated with using AI, makes it possible to mobilise employers to take specific measures to protect employees' mental and physical health.
Artificial Intelligence for data-driven decision-making and governance in public affairs
2022, Government Information Quarterly
Literature shows there is a growing interest in studies involving Artificial Intelligence (AI) in the public sector; and while there is evidence of many governmental initiatives that have been established to harness the power of AI, empirical research on the topic and evidence-based insights are rather lacking. The aim of this Special Issue on Artificial Intelligence for Data-Driven Decision-Making and Governance in Public Affairs is to extend both the theoretical and practical boundaries of AI research in the public sector in order to improve governmental decision-making and governance, thus enhancing public value creation. The papers in this special issue focus on AI risks and guidelines, AI governance, the risks of governmental implementation of AI to citizens' privacy, increasing citizen satisfaction through AI-enabled government services, the enablers and challenges of AI implementation in specific public sectors, and using AI to study political opinion. These papers not only advance our knowledge and understanding of the use of AI in government and public governance, but they also help to set out a renewed research agenda. Future research should, among other things, focus on inter- and multi-disciplinary empirical studies that call for the collaboration of a variety of stakeholders; on the longitudinal dynamics of creating public value through the breadth and depth of AI assimilation; and on the investigation of the ethical challenges (particularly data privacy) in AI implementation.
Harnessing the Potential of Artificial Intelligence to Foster Citizens’ Satisfaction: An empirical study on India
2022, Government Information Quarterly
Citation Excerpt :
The assimilation gap is attributed to the fact that actual use often lags behind the decision to adopt AI in government sectors (Chatfield & Reddick, 2018). Further, since AI substitutes for humans, the application of AI services is associated with risks (Butterworth, 2018; Čerka, Grigienė, & Sirbikytė, 2017), which can impact decision-making processes and, in some instances, have a detrimental effect on implementation of AI services. Major implementation issues include principles of explicability, beneficence, non-maleficence, and justice, together with privacy and security (Floridi et al., 2018).
Governments are increasingly employing artificial intelligence (AI) enabled services though this is still a relatively new concept that is in nascent stages of implementation. Despite growing emphasis by governments on employing AI-enabled services, many citizens are skeptical of their benefits; this makes an analysis of AI-enabled services an important area of research, especially from the perspective of citizens. This paper employs IT assimilation theory and public value theory to develop a theoretical model that examines whether the introduction of AI-enabled services would generate public value for citizens in India. The model employs the Partial Least Square-Structural Equation Modeling (PLS-SEM) technique to examine how risk factors impact the uptake of AI-enabled services in India. Based on 315 interviews conducted in India, the study highlights that the breadth and depth assimilation of AI-enabled services positively impacts and enhances the satisfaction of citizens, which in turn generates public value.
Orchestrating artificial intelligence for urban sustainability
2022, Government Information Quarterly
Citation Excerpt :
On the other hand, the utilization of AI techniques helps to make sense of big data (Yaqoob et al., 2016). Being able to process and link data in unprecedented ways and at superfast speed (Chatterjee, Kar, & Gupta, 2017; Newell & Marabelli, 2015), AI systems can operate autonomously, even without any human intervention, to learn and identify patterns in big data for decision-making (Janssen et al., 2020) and to draw targeted conclusions about specific situations (Čerka, Grigienė, & Sirbikytė, 2017). AI provides several benefits, such as making better decisions, optimizing internal operations, improving the quality of products and services, freeing up humans for more creative work, facilitating innovation, and pursuing new markets (Alshahrani, Dennehy, & Mäntymäki, 2021;Davenport & Ronanki, 2018; Duan et al., 2019).
Artificial intelligence (AI) is regarded as the next digital frontier in government, with many potential applications for economic development as well as sustainable urbanization. Governments have started experimenting with AI, but empirical research on how to leverage and implement AI remains limited. This study analyzed two cases of AI implementation in a large city and identified various AI capabilities useful for government. More importantly, purposeful orchestration of AI-related resources such as data, knowledge, algorithms, and information systems is necessary for developing strong AI capabilities. The findings indicate two different types of orchestration: policy-driven orchestration focuses on the integration of resources, while innovation-driven orchestration focuses on triangulation. This study contributes to the growing body of knowledge on AI in government by revealing and conceptualizing different paths and approaches to AI implementation. They also serve to inform practitioners' planning of AI implementation.
The future of AI politics, policy, and business
2024, Business and Politics
Data-Driven Decision Support Systems in E-Governance: Leveraging AI for Policymaking
2024, Lecture Notes in Networks and Systems

View all citing articles on Scopus

Paulius Čerka, Attorney in Čerka and The Partners advocates bureau, Associate Professor; Dr., Vytautas Magnus University Faculty of Law, Department of Private Law, Kaunas, Lithuania.

Jurgita Grigienė, Corresponding author: Attorney in Čerka and The Partners advocates bureau, Associate Professor; Dr., Vytautas Magnus University Faculty of Law, Department of Private Law, Kaunas, Lithuania.

Gintarė Sirbikytė, Attorney in Čerka and The Partners Advocates Bureau, European lawyer Registered by Solicitors Regulation Authority.

View full text