Three Laws Blog: Law and Artificial Intelligence

Brennan on Proposals for the Rights and Liabilities of Artificial Intelligence through Electronic Personhood: Conscious Machines, Hermit Crabs, and the Distant Past

Dan Brennan has published Proposals for the Rights and Liabilities of Artificial Intelligence through Electronic Personhood: Conscious Machines, Hermit Crabs, and the Distant Past. Here is the abstract.

Examines the complexities of contemporary AI regulation, chiefly in the technology's unique capacity for autonomous decision making and independent learning. The paper highlights the potential of granting legal personhood to sophisticated AI systems as an alternative to current regulatory schemes. Three categories of personhood are considered: the ontological model, corporate frameworks, and the ancient Roman concept of peculium. Fundamentally, each model allows AI to hold property and be responsible for its actions, notably by indemnifying any legal persons it may harm. That said, each model has unique benefits and limitations which are discussed in turn. Chapter 6 concludes that neither one of these models can be exclusively relied upon. Instead, a new form of 'electronic personhood' should adopt elements from all the regulatory methods discussed herein, including methods of current regulation and the EU's AI Act.

Download the article from SSRN at the link.

Bilokon on Evolution-Boostrapped Simulation: Artificial or Human Intelligence: Which Came First? @bilokon @imperialcollege

Paul Bilokon, Imperial College Lndon, has published Evolution-Bootstrapped Simulation: Artificial or Human Intelligence: Which Came First? Here is the abstract.

Humans have created artificial intelligence (AI), not the other way around. This statement is deceptively obvious. In this note, we decided to challenge this statement as a small, lighthearted Gedankenexperiment. We ask a simple question: in a world driven by evolution by natural selection, would neural networks or humans be likely to evolve first? We compare the Solomonoff--Kolmogorov--Chaitin complexity of the two and find neural networks (even LLMs) to be significantly simpler than humans. Further, we claim that it is unnecessary for any complex human-made equipment to exist for there to be neural networks. Neural networks may have evolved as naturally occurring objects before humans did as a form of chemical reaction-based or enzyme-based DNA computation. Now that we know that neural networks can pass the Turing test and suspect that they may be capable of superintelligence, we ask whether the natural evolution of neural networks could lead from pure evolution by natural selection to what we call evolution-bootstrapped simulation. The evolution of neural networks does not involve irreducible complexity; would easily allow irreducible complexity to exist in the evolution-bootstrapped simulation; is a falsifiable scientific hypothesis; and is independent of / orthogonal to the issue of intelligent design.

Download the essay from SSRN at the link.

Osborne on Personhood for Synthetic Beings: Legal Parameters and Consequences of the Dawn of Humanlike AI @SCHTLJ

Daniel Osborne has published Personhood for Synthetic Beings: Legal Parameters and Consequences of the Dawn of Humanlike Artificial Intelligence at 37 Santa Clara High Technology Law Journal 257 (2021). Here is the abstract.

There is a growing interest in Artificial Intelligence in tech hubs like Silicon Valley and in the world at large that is finally matching the ambition and creativity that movies and media have been producing for a very long time. This paper takes that cultural interest and economic investment to its logical conclusion and assumes that a humanlike Artificial Intelligence, or synthetic person, can and will be created in the near future. Then it considers what the law should do about that creation. Specifically, this paper argues that a test should be developed for whether a synthetic person can be granted personhood under the law before such test is needed. The tenets, if not the specifics, of such a test are outlined here that provide for a synthetic person that is intelligent, social, self-conscious, and individualized. But that is only half the problem. If a synthetic person can be created, one that passes a pre-determined test that allows it to access legal standing, to what sorts of rights and responsibilities does it have access? In light of the kind of synthetic person that could pass the test, four potential outcomes for rights for a synthetic person are addressed ranging from none to a full suite of human civil rights.

Download the article from SSRN at the link.

Schroeder on Robots and Rights: Reviewing Recent Positions in Legal Philosophy and Ethics @Faust_III

Wolfgang M. Schroeder, University of Wuerzberg, has published Robots and Rights: Reviewing Recent Positions.  Here is the abstract.

Controversies about the moral and legal status of robots and of humanoid robots in particular are among the top debates in recent practical philosophy and legal theory. As robots become increasingly sophisticated, and engineers make them combine properties of tools with seemingly psychological capacities that were thought to be reserved for humans, such considerations become pressing. While some are inclined to view humanoid robots as more than just tools, discussions are dominated by a clear divide: What some find appealing, others deem appalling, i.e. “robot rights” and “legal personhood” for AI systems. Obviously, we need to organize human–robot interactions according to ethical and juridical principles that optimize benefit and minimize mutual harm. Avoiding disrespectful treatment of robots can help to preserve a normative basic ethical continuum in the behaviour of humans. This insight can contribute to inspire an “overlapping consensus” as conceptualized by John Rawls in further discussions on responsibly coordinating human/robot interactions.

Download the article from SSRN at the link.

Loza de Siles on AI, on the Law of Being

Emile Loza de Siles, Duquesne University Schools of Law; Technology & Cybersecurity Law Group; University of Maryland Global Campus; has published AI, on the Law of Being: “Feminine” Imagery in Humanoid Robots, Evolving Law as to What Constitutes a Person [Abstract] as Duquesne University School of Law Research Paper No. 2020-12. Here is the abstract.

Mechanized embodiments of artificial intelligence (“AI”) in robots may take humanoid form. Images of agile, tall, and broad-shouldered robots bring to mind untold ranks of lethal autonomous soldiers.” Robots in forms slope-haunched, big-chested, and stalk-gaited call forward images of ferocious dogs used to track and attack troops in World War II or, seen in television broadcasts, of protestors during the Vietnam War or the Civil Rights movement. Perhaps the world’s most famous humanoid robot is “Sophia.” A remarkable machine learning, voice-synthesizing robot, Sophia’s humanoid form uses feminine and pseudonormative imagery: “Caucasian” pale “skin,” blue “eyes” below arched eyebrows, and a notable “breast line” atop a slim-hipped feminine-appearing body. Impeccably styled in designer clothing, the robot has graced the covers of ELLE, COSMOPOLITAN, and other magazines. The robot has operated before the United Nations and so impressed the Saudi Arabian government that it granted this “feminine” robot a first-in-world national citizenship, an unimaginable irony where liberties are sharply curtailed for women, who are often legally no more than chattel. Sophia makers now have built a prepubescently femininely-imaged and evocative “Little Sophia” robot. In promotional media, a girl child features prominently in interactions with the Little Sophia robot, which, among others, incorporates image capture and facial recognition capabilities behind its piquantly lashed, exaggeratedly large doe eyes. What does such “feminine” imagery for these robots connote? What role does it play? What is its calculated purpose? Is it meant to evoke stereotypes of feminine subservience? Does the “feminine” imagery of the Sophia and Little Sophia allow humans will feel less threatened by robots and to anthropomorphize them more easily? Does this, in turn, aim to accelerate the social acceptance of robots as part of daily life and, again calling on feminine stereotypes, trust them more readily? Research and anecdote suggest this is so. As with feminine-appearing sex dolls and robots in Japan and elsewhere, does this imagery lead some men to embrace mechanosexuality and to prefer humanoid “females” over women? Reports from Japan suggest this is also so, as do $10,000 bids from men to have sex with a “virgin” robot. What are the legal evolutionary impacts of the use of feminine imagery for robots and the triggering by that imagery of female stereotypes? Does it portend a denigration of what it means to be a person under the law? Specifically, does it degrade how the law views and treats a female human person? If, for example, rape of a robot person is a legal impossibility, what devolution of law may that threaten for a woman, a girl? Saudi citizenship for the Sophia robot and European legal developments toward robot citizenship foretell that the realm of legal personhood is expanding to include inanimate, but mechanically embodied, intelligence-mimicking things. What does this portend for humans? The attachment of legal personhood to corporations, for example, has been a workable, if not always just, legal fiction. Citizen United, however, shows what democracy-devastating consequences of that fiction. What greater harm could result from including robots and other AI systems within the legal scope of “person”? If a robot is a person, would the infamy of Dred Scott drive courts to conclude that a robot is likewise a citizen? If a robot is a person, what about Her? Are there countervailing benefits to attaching personhood and citizenship to robots or other AI instantiations? What about permitting the use of evocative imagery for robots, such as for in-home robots that function as medication reminder and tracking systems, and de facto companions, to seniors living alone? What about robots that help children with autism spectrum conditions learn to become more socially skilled? This work explores these pressing questions and creates opportunity to formulate others.

Mik on AI as a Legal Person?

Eliza Mik Melbourne Law School; Singapore Management University; TILT; is publishing AI as a Legal Person? in Artificial Intelligence & Intellectual Property (Reto Hilty & Kung-Chung Li, eds., OUP, 2020) (Forthcoming).

Cyclical advancements in artificial intelligence (“AI”) are usually accompanied by theories advocating the granting of legal personhood to sophisticated, autonomous computers. The paper criticizes such theories as incorrect — a possible result of legal scholars being seduced by incomprehensible technical terminology, sensationalistic stories in the popular press and “creative” photo filters that transform our faces into animals. Discussions as to when computers should be recognized as persons are, logically, outside of the scope of intellectual property law. The granting of legal personhood is not premised on the existence of consciousness, intelligence or creativity. Recognizing an entity as a legal person is a normative choice dictated by commercial expediency, not the result of fulfilling any technical criteria. While it is necessary to acknowledge the blurring of borders between art and (computer) science as well as the increase in the technological sophistication of the tools used by authors and inventors, it is also necessary to state that even an exponential increase in “computer creativity” will not sever the link between the computer and its user. Before discarding the idea of legal personhood for “creative algorithms” once and for all, the paper explores the relationships between autonomy and creativity. In particular, it places technical terms such as “AI” and “autonomy” in their original context and criticizes uninformed attempts to imbue them with normative connotations.

Download the essay from SSRN at the link.

Banteka on Artificially Intelligent Persons @nbanteka

Nadia Banteka, Villanova University School of Law, is publishing Artificially Intelligent Persons in the Houston Law Review (volume 58) (2020). Here is the abstract.

Artificial Intelligence (AI) entities seriously challenge traditional legal frameworks for attribution and liability because they operate at an increasing distance from their developers and owners, resulting in accountability gaps. Consider a scenario in which a self-driving car causes injury or even death to a human. Who do we hold accountable? We have no clear answer as to who can be sued or prosecuted because we lack a comprehensive legal understanding of AI entities. Many scholars propose as a solution to the accountability problem attaching liability to the direct source of the harm, the AI entity itself, by first granting it legal personhood. But the law has yet to answer the question of whether AI entities qualify for legal personhood and, if so, on what legal basis. This article is the first to empirically assess the scope of legal personhood as it relates to AI entities, and to answer this question. I make two claims about the problem of legal personhood for AI. First, I argue that the courts’ overall approach to legal personhood has been more disparate than many have assumed, and it does not support legal personhood for AI entities. To substantiate this position, I analyze the legal basis for judicial decisions conferr ing legal personhood on artificial entities across U.S. courts from 1809 to the present, and I offer a statistical analysis of the frequency with which different conditions for legal personhood appear in these decisions. I find a clear dissonance between legal doctrine and existing theory on legal personhood for AI entities. Second, I argue that empirically understanding the legal landscape for legal personhood prevents courts from conferring legal personhood on AI entities, and should give legislators pause before doing so. If courts and legislators consider the conditions for legal personhood that this article identifies in answering questions of liability for AI entities, they will discover the incompatibility between legal personhood and these entities. Without recognition of this incompatibility, theory, policy, and litigation surrounding AI entities could move in a direction that undermines legal certainty and upsets legal expectation.

Download the article from SSRN at the link.B

Lior on AI Entities as AI Agents @AnatLior @yaleisp

Anat Lior, Yale University Law School; Yale Information Society Project, is publishing AI Entities as AI Agents: Artificial Intelligence Liability and the AI Respondeat Superior Analogy in volume 46 of the Mitchell Hamline Law Review (2020). Here is the abstract.

Artificial Intelligence (AI) based entities are already causing damages and fatalities in today's commercial world. As a result, the dispute about tort liability of AI-based machines, algorithms, agents and robots is exponentially advancing in the scholarly world and outside of it. When it comes to AI accidents, different scholars and key figures in the AI industry advocate for different liability regimes. This ever-growing disagreement is condemning this new emergent technology, soon to be found in almost every home and street in the US and around the world, into a realm of regulatory uncertainty. This obstructs our ability to fully enjoy the many benefits AI has to offer us as consumers and as a society. This Article advocates for the adoption and application of a strict liability regime on current and future AI accidents. It does so by delving into and exploring the realm of legal analogies in the AI context and promoting the agency analogy, and subsequently, the respondeat superior doctrine. This Article explains and justifies why the agency analogy is the best-suited one in contrast to other analogies which have been suggested in the context of AI liability (e.g., products, animals, electronic persons and even slaves). As a result, the intuitive application of the respondeat superior doctrine provides the AI industry with a much-needed underlying liability regime which will enable it to continue to evolve in the years to come, and its victim to receive remedy once accidents occur.

Download the article from SSRN at the link.

Chesterman on We, the Robots @ProfChesterman

Simon Chesterman, National University of Singapore, has published We, the Robots?. Here is the abstract.

As computers surpass human intelligence and take on greater responsibilities, should they be given rights also? This essay considers the evolution of the Turing Test for machine intelligence and its relationship to ongoing debates about legal personality and artificial intelligence.

Download the essay from SSRN at the link.

Jaynes on Legal Personhood for AI

Tyler L. Jaynes, Utah Valley University, has published Legal Personhood for Artificial Intelligence: Citizenship as the Exception to the Rule, at 34 AI & Society 1 (2019). Here is the abstract.

The concept of artificial intelligence is not new nor is the notion that it should be granted legal protections given its influence on human activity. What is new, on a relative scale, is the notion that artificial intelligence can possess citizenship—a concept reserved only for humans, as it presupposes the idea of possessing civil duties and protections. Where there are several decades’ worth of writing on the concept of the legal status of computational artificial artefacts in the USA and elsewhere, it is surprising that law makers internationally have come to a standstill to protect our silicon brainchildren. In this essay, it will be assumed that future artificial entities, such as Sophia the Robot, will be granted citizenship on an international scale. With this assumption, an analysis of rights will be made with respect to the needs of a non-biological intelligence possessing legal and civic duties akin to those possessed by humanity today. This essay does not present a full set of rights for artificial intelligence—instead, it aims to provide international jurisprudence evidence aliunde ab extra de lege lata for any future measures made to protect non-biological intelligence.

Download the article from the website here.