We are sceptical of concerns over the opacity of algorithmic decision tools. While transparency and explainability are certainly important desiderata in algorithmic governance, we worry that automated decision-making is being held to an unrealistically high standard, possibly owing to an unrealistically high estimate of the degree of transparency attainable from human decision-makers. In this paper, we review evidence demonstrating that much human decision-making is fraught with transparency problems, show in what respects AI fares little worse or better and argue that (...) at least some regulatory proposals for explainable AI could end up setting the bar higher than is necessary or indeed helpful. The demands of practical reason require the justification of action to be pitched at the level of practical reason. Decision tools that support or supplant practical reasoning should not be expected to aim higher than this. We cast this desideratum in terms of Daniel Dennett’s theory of the “intentional stance” and argue that since the justification of action for human purposes takes the form of intentional stance explanation, the justification of algorithmic decisions should take the same form. In practice, this means that the sorts of explanations for algorithmic decisions that are analogous to intentional stance explanations should be preferred over ones that aim at the architectural innards of a decision tool. (shrink)

The danger of human operators devolving responsibility to machines and failing to detect cases where they fail has been recognised for many years by industrial psychologists and engineers studying the human operators of complex machines. We call it “the control problem”, understood as the tendency of the human within a human–machine control loop to become complacent, over-reliant or unduly diffident when faced with the outputs of a reliable autonomous system. While the control problem has been investigated for some time, up (...) to this point its manifestation in machine learning contexts has not received serious attention. This paper aims to fill that gap. We argue that, except in certain special circumstances, algorithmic decision tools should not be used in high-stakes or safety-critical decisions unless the systems concerned are significantly “better than human” in the relevant domain or subdomain of decision-making. More concretely, we recommend three strategies to address the control problem, the most promising of which involves a complementary coupling between highly proficient algorithmic tools and human agents working alongside one another. We also identify six key principles which all such human–machine systems should reflect in their design. These can serve as a framework both for assessing the viability of any such human–machine system as well as guiding the design and implementation of such systems generally. (shrink)

A concise but informative overview of AI ethics and policy. -/- Artificial intelligence, or AI for short, has generated a staggering amount of hype in the past several years. Is it the game-changer it's been cracked up to be? If so, how is it changing the game? How is it likely to affect us as customers, tenants, aspiring homeowners, students, educators, patients, clients, prison inmates, members of ethnic and sexual minorities, and voters in liberal democracies? Authored by experts in fields (...) ranging from computer science and law to philosophy and cognitive science, this book offers a concise overview of moral, political, legal and economic implications of AI. It covers the basics of AI's latest permutation, machine learning, and considers issues such as transparency, bias, liability, privacy, and regulation. -/- Both business and government have integrated algorithmic decision support systems into their daily operations, and the book explores the implications for our lives as citizens. For example, do we take it on faith that a machine knows best in approving a patient's health insurance claim or a defendant's request for bail? What is the potential for manipulation by targeted political ads? How can the processes behind these technically sophisticated tools ever be transparent? The book discusses such issues as statistical definitions of fairness, legal and moral responsibility, the role of humans in machine learning decision systems, “nudging” algorithms and anonymized data, the effect of automation on the workplace, and AI as both regulatory tool and target. (shrink)

In this article, I argue that lawmakers must abandon their previous reluctance to engage with questions of personal identity (PI). While frequently seen as an esoteric subject, of limited interest outside of academic philosophy departments, I attempt to show that, in fact, assumptions about PI—and its durability in the face of certain psychological or genetic changes—underpin many current legal rules. This is most perhaps obviously exemplified with regard to reproductive technologies. Yet the Parfitian challenge to identify a victim of ‘bad’ (...) reproductive choices has been largely overlooked in framing legislative responses to such technologies. Furthermore, I argue, it is not only with regard to emerging technologies that questionable assumptions about PI play a role; legal responses to questions about the attribution of criminal responsibility, and about the treatment of demented or brain-injured people, necessitate a frank engagement with such questions. It may be, however, that a multi-faceted approach to PI, which takes account of genetic, psychological and social factors—will prove a better fit for the myriad needs of the legal system than any sort of ‘unified theory of identity’. (shrink)

Robert Nozick, in what is surely one of the most intriguing and provocative footnotes in modern philosophical writing, referred in Anarchy,StateandUtopia to the notion of a In keeping with the central arguments of that text, his suggestion was that choices about the genetic composition of future generations should, as far as possible, be left in the hands of private individuals, and should not be determined or restricted by the state. This free market in genetic screening would meet and would possess (...) In short, prospective parents would be allowed, to whatever extent was rendered possible by current technology, to choose the genetic traits of their future children. (shrink)

Artificial Intelligence (AI) is a diverse technology. It is already having significant effects on many jobs and sectors of the economy and over the next ten to twenty years it will drive profound changes in the way New Zealanders live and work. Within the workplace AI will have three dominant effects. This report (funded by the New Zealand Law Foundation) addresses: Chapter 1 Defining the Technology of Interest; Chapter 2 The changing nature and value of work; Chapter 3 AI and (...) the employment relationship; Chapter 4 Consumers, professions and society. The report includes recommendations to the New Zealand Government. (shrink)

Lord Scarman’s judgment about when someone under the age of 16 years should have the right to make their own medical decisions emphasised the decision-making abilities of the particular child. He said: > …the parental right to determine whether or not their minor child below the age of 16 will have medical treatment terminates if and when the child achieves a sufficient understanding and intelligence to enable him or her to understand fully what is proposed.1 That created a duty on (...) healthcare practitioners to assess whether a particular minor has decision-making abilities at a degree that would enable them to understand the decision to a high extent, sufficient hopefully that they would ‘own’ the decision. In December of 2020, the High Court considered whether young people with gender dysphoria and seeking access to puberty blocking therapy, were likely to pass Scarman’s mature minor test and cast doubt on their ability to fully understand that decision, thereby making it less likely that a healthcare practitioner would decide they are a mature minor for that therapy. The High Court said: > It is highly unlikely that a child aged 13 or under would be competent to give consent to the administration of puberty blockers. It is doubtful that a child …. (shrink)