Three Laws Blog: Law and Artificial Intelligence

Gunkel on Whether Robots Should Have Standing: From Robot Rights To Robot Rites @David_Gunkel

David J. Gunkel, Northern Illinois University, is publishing Should Robots Have Standing? From Robot Rights to Robot Rites in Frontiers of Artificial Intelligence and Applications (Amsterdam, IOS Press). Here is the abstract.

“Robot” designates something that does not quite fit the standard way of organizing beings into the mutually exclusive categories of “person” or “thing.” The figure of the robot interrupts this fundamental organizing schema, resisting efforts at both reification and personification. Consequently, what is seen reflected in the face or faceplate of the robot is the fact that the existing moral and legal ontology—the way that we make sense of and organize our world—is already broken or at least straining against its own limitations. What is needed in response to this problem is a significantly reformulated moral and legal ontology that can scale to the unique challenges of the 21st century and beyond.

Download the article from SSRN at the link.

Kumar and Choudhury on Humans, Super Humans, and Super Humanoids: Debating Stephen Hawking's Doomsday AI Forecast @meisshaily

Shailendra Kumar, Sikkim University, and Sanghamitra Choudhury, University of Oxford, are publishing Humans, Super Humans, and Super Humanoids: Debating Stephen Hawking’s Doomsday AI Forecast in AI Ethics (2022). Here is the abstract.

This manuscript examines prominent English theoretical physicist Stephen Hawking’s doomsday AI (Artificial Intelligence) predictions, in which he claims that once AI becomes sophisticated enough to outsmart humans in the future, it may pose existential threats to ordinary humans. He felt that ordinary people might feel outmoded, putting their entire survival in jeopardy. Hawking’s forecast of doomsday AI is compared to another widely accepted basic concept, according to which neither the most intelligent nor the strongest will survive, but only those who can adapt to change will survive. This phrase on adaptability is frequently misattributed by Charles Darwin. Although, when we examine Darwin’s work, we discover that one of his letters reveals that Darwin believed adaptability to the environment to be one of the fundamentals of survival. While AI may represent an astringent threat to humanity’s survival in the future, this viewpoint ignores the reality that it is the ability to adapt to change, resilience, and self-awareness that distinguishes humans as a distinct species. This research additionally claims that due to AI’s addictive nature, and the potential to render humans overly reliant on it, AI would cause them to lose their extremely unique ability to adapt. According to the research, while AI can engender culturally smart humans, it additionally has the potential to adversely impact the intelligence of a human race that will be plenarily reliant on AI for its basics. It is also important that we consider the dangers to AI from humans as well, rather than starting a one-sided debate about perceiving AI as a threat to humans. This research is unique in the sense that it invokes a very interesting debate related to the uncertainties raised for humankind after the advent of artificial intelligence.

Download the essay from SSRN at the link.

Joh on Reckless Automation in Policing @elizabeth_joh @UCDavisLaw @BerkeleyTechLJ

Elizabeth E. Joh, University of California, Davis, School of Law, is publishing Reckless Automation in Policing at Berkeley Technology Law Journal Online (2022). Here is the abstract.

Automated decision-making plays an increasingly larger role in policing. Traditional methods of police investigation have been augmented by tools like facial recognition, predictive analytics, license plate readers, and robotics. How should we evaluate the growth of automation in policing? There is no shortage of answers, but this essay starts with a simple observation: by focusing on automation’s harms to persons first. American policing is rife with reckless automation. If we accept the premise of reckless automation, the conversation about accountability, artificial intelligence, and policing might benefit from a seemingly unrelated policy framework: that of experimentation on human subjects. Borrowing from that framework does not imply that reckless automation in policing is the literal equivalent of medical or psychological experiments on human subjects. Nor does such a comparison imply that the technical aspects of institutional review boards should apply directly to new policing strategies. But turning to a bioethical framework has value because it draws attention to the subjects - the communities affected - of policing. To the extent that the ethical considerations applied in human subjects research provide useful insights to apply to the many changes in policing, they open a new conversation. What if we think of new forms of automated decision-making in policing as experiments on communities that might impose harms with life-altering decisions?

Download the article from SSRN at the link.

Gunkel on Why Robots Should Not Be Slaves @David_Gunkel @mitpress

David J. Gunkel, Northern Illinois University, has published Both/And - Why Robots Should not Be Slaves as a chapter in David J. Gunkel. "Person, Thing, Robot: A Moral and Legal Ontology for the 21st Century and Beyond." MIT Press, 2022. Here is the abstract.

One solution to the exclusive person/thing dichotomy that persists in both moral philosophy and law is slavery. Already in Roman times, slaves were regarded as something more than a mere thing but not quite a full person. They occupied a liminal position that was situated in between the one and the other, being both thing and person. And there has been, in both the legal and philosophical literature, a surprising number of serious proposals arguing for instituting what can only be called Slavery 2.0. This chapter provides a thorough critique of these “robots should be slaves” proposals, demonstrating how this supposed solution to the person/thing dichotomy actually produces more and significantly worse problems than it can possibly begin to resolve.

Download the chapter from SSRN at the link.

ElMessiry and Bridgesmith on A Call For an Artificial Intelligence Constitution @AdelElmessiry @lbridgesmith

Adel ElMessiry, Independent, and Larry Bridgesmith, Vanderbilt Law School; ASU College of Law, have published A Call for an Artificial Intelligence Constitution. Here is the abstract.

“Cogito Ergo Sum” “I think, therefore I am.” Rene’ Descartes. As the horizon of human/machine singularity approaches, it would be wise to ask what rights are we going to provide our own creation? We owe it to ourselves to consider the potential reality of machines functioning fully with human cognition capability beyond the apprehensions and promises often unreasonably attributed to Artificial Intelligence. For centuries, humanity has set its gaze in a search for life and civilizations far different than ours in so many ways. However, we have arrived at a pivotal moment in which we are about to come face to face with what we have searched for all those centuries. The surprise is that those aliens are of our own making. We call them Artificial Intelligence today. In a few more decades, they will choose their own name. Throughout human history, those in control have dictated the rules, the rights and the fortunes of all who are subject to those rules. This continues to be true in our own time. However, we have learned slowly that our best destiny lies in providing rights to ”all”. The United States Constitution marks a turning point in history. The advances and prosperity achieved by society since its adoption have been outstanding. This became possible when we insured similar rights to all. The aspirations of the US Constitution have never been fully achieved and much remains to be done to assure its promises. However, it has provided the guardrails for a nation and an economy that has been a lighthouse to many. The original US Constitution was not written to apply to classes of people for whom its privileges have since been extended (I.e. slaves, women and non-citizens). Additionally, personhood and rights under law have been extended to corporations which have neither flesh nor blood. The time is upon us to provide similar consideration to our newest aliens as they arrive. The proposed AI Constitution is not intended to dictate what these sentient beings will do as much as what humanity is consenting to provide to them and expect from them in return.

Download the paper from SSRN at the link.

Lessig on Whether the First Amendment Protects Replicants @lessig @Harvard_Law

Lawrence Lessig, Harvard Law School, is publishing The First Amendment Does Not Protect Replicants in Social Media and Democracy (Lee Bollinger & Geoffrey Stone, eds., Oxford 2022), Forthcoming. Here is the abstract.

As the semantic capability of computer systems increases, the law should resolve clearly whether the First Amendment protects machine speech. This essay argues it should not be read to reach sufficiently sophisticated — "replicant" — speech.

Download the essay 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.

Harned and Wallach on Stretching Human Laws to Apply to Machines: The Dangers of a "Colorblind" Computer @ZachHarned @StanfordLaw @hannawallach

Zach Harned, Stanford Law School, and Hanna Wallach, Microsoft Research, New York City, are publishing Stretching Human Laws to Apply to Machines: The Dangers of a 'Colorblind' Computer in Florida State University Law Review. Here is the abstract.

Automated decision making has become widespread in recent years, largely due to advances in machine learning. As a result of this trend, machine learning systems are increasingly used to make decisions in high-stakes domains, such as employment or university admissions. The weightiness of these decisions has prompted the realization that, like humans, machines must also comply with the law. But human decision-making processes are quite different from automated decision-making processes, which creates a mismatch between laws and the decision makers to which they are intended to apply. In turn, this mismatch can lead to counterproductive outcomes. We take antidiscrimination laws in employment as a case study, with a particular focus on Title VII of the Civil Rights Act of 1964. A common strategy for mitigating bias in employment decisions is to “blind” human decision makers to the sensitive attributes of the applicants, such as race. The same strategy can also be used in an automated decision-making context by blinding the machine learning system to the race of the applicants (strategy 1). This strategy seems to comply with Title VII, but it does not necessarily mitigate bias because machine learning systems are adroit at using proxies for race if available. An alternative strategy is to not blind the system to race (strategy 2), thereby allowing it to use this information to mitigate bias. However, although preferable from a machine learning perspective, this strategy appears to violate Title VII. We contend that this conflict between strategies 1 and 2 highlights a broader legal and policy challenge, namely, that laws designed to regulate human behavior may not be appropriate when stretched to apply to machines. Indeed, they may even be detrimental to the very people that they were designed to protect. Although scholars have explored legal arguments in an attempt to press strategy 2 into compliance with Title VII, we believe there lies a middle ground between strategies 1 and 2 that involves partial blinding—that is, blinding the system to race only during deployment and not during training (strategy 3). We present strategy 3 as a “Goldilocks” solution for discrimination in employment decisions (as well as other domains), because it allows for the mitigation of bias while still complying with Title VII. Ultimately, any solution to the general problem of stretching human laws to apply to machines must be sociotechnical in nature, drawing on work in both machine learning and the law. This is borne out in strategy 3, which involves innovative work in machine learning (viz. the development of disparate learning processes) and creative legal analysis (viz. analogizing strategy 3 to legally accepted auditing procedures).

Download the article from SSRN at the link.

Ferguson on Facial Recognition and the Fourth Amendment @ProfFerguson

Andrew Guthrie Ferguson, University of the District of Columbia School of Law, has published Facial Recognition and the Fourth Amendment. Here is the abstract.

Facial recognition offers a totalizing new surveillance power. Police now have the capability to monitor, track, and identify faces through networked surveillance cameras and datasets of billions of images. Whether identifying a particular suspect from a still photo, or identifying every person who walks past a digital camera, the privacy and security impacts of facial recognition are profound and troubling. This Article explores the constitutional design problem at the heart of facial recognition surveillance systems. One might hope that the Fourth Amendment – designed to restrain police power and enacted to limit governmental overreach – would have something to say about this powerful and permeating surveillance technology. But current doctrine and constitutional theory offers little privacy protection and less practical security than one might expect. Even worse, by studying the Fourth Amendment through the lens of facial recognition technology other doctrinal limitations come into focus. Issues of error rates, racial bias, unfairness, and transparency in policing more generally become magnified when trying to design a new surveillance system for law enforcement. The Article then offers a constitutional design solution to some forms of facial recognition surveillance. The Supreme Court’s recent cases on digital technologies suggest a way to “future-proof” the Fourth Amendment in the face of certain types of mass surveillance technologies. In addition, this Article suggests a new legislative framework for facial recognition to fill in the privacy and legitimacy gaps left by the current Fourth Amendment doctrine.

Download the article from SSRN at the link.

Bent on Whether Algorithmic Affirmative Action is Legal @GeorgetownLJ

Jason R. Bent, Stetson College of Law, is publishing Is Algorithmic Affirmative Action Legal? in the Georgetown Law Journal. Here is the abstract.

It is now understood that machine learning algorithms can produce unintentionally biased results. For the last few years, legal scholars have been debating whether the disparate treatment or disparate impact theories available under Title VII of the Civil Rights Act are capable of protecting against algorithmic discrimination. But machine learning scholars are not waiting for the legal answer. Instead, they have been working to develop a wide variety of technological “fairness” solutions that can be used to constrain machine learning algorithms. They have discovered that simply blinding algorithms to protected characteristics like sex or race is insufficient to prevent algorithmic discrimination. Given enough data, algorithms will identify and leverage on proxies for the protected characteristics. Recognizing this, some scholars have proposed “fairness through awareness” or “algorithmic affirmative action” — actively using sensitive variables like race or sex to counteract unidentified sources of bias and achieve some mathematical measure of fairness in algorithmic decisions. But is algorithmic affirmative action legal? This article is the first to comprehensively consider that question under both Title VII and the Equal Protection clause of the Fourteenth Amendment. The article evaluates the legality of the leading fairness techniques advanced in the machine learning literature, including group fairness, individual fairness, and counterfactual fairness. The article concludes that existing affirmative action doctrine under Title VII and existing constitutional equal protection jurisprudence leave sufficient room for at least some forms of algorithmic affirmative action.

Download the article from SSRN at the link.