It has been argued that implicit biases are operative in philosophy and lead to significant epistemic costs in the field. Philosophers working on this issue have focussed mainly on implicit gender and race biases. They have overlooked ideological bias, which targets political orientations. Psychologists have found ideological bias in their field and have argued that it has negative epistemic effects on scientific research. I relate this debate to the field of philosophy and argue that if, as some studies (...) suggest, the same bias also exists in philosophy then it will lead to hitherto unrecognised epistemic hazards in the field. Furthermore, the bias is epistemically different from the more familiar biases in respects that are important for epistemology, ethics, and metaphilosophy. (shrink)

In this paper, we distinguish between different sorts of assessments of algorithmic systems, describe our process of assessing such systems for ethical risk, and share some key challenges and lessons for future algorithm assessments and audits. Given the distinctive nature and function of a third-party audit, and the uncertain and shifting regulatory landscape, we suggest that second-party assessments are currently the primary mechanisms for analyzing the social impacts of systems that incorporate artificial intelligence. We then discuss two kinds of (...) as-sessments: an ethical risk assessment and a narrower, technical algo-rithmic bias assessment. We explain how the two assessments depend on each other, highlight the importance of situating the algorithm within its particular socio-technical context, and discuss a number of lessons and challenges for algorithm assessments and, potentially, for algorithm audits. The discussion builds on our team’s experience of advising and conducting ethical risk assessments for clients across dif-ferent industries in the last four years. Our main goal is to reflect on the key factors that are potentially ethically relevant in the use of algo-rithms, and draw lessons for the nascent algorithm assessment and audit industry, in the hope of helping all parties minimize the risk of harm from their use. (shrink)

Future-bias is the preference, all else being equal, for negatively valenced events be located in the past rather than the future, and positively valenced ones to be located in the future rather than the past. Strong risk aversion is the preference to pay some cost to mitigate the badness of the worst outcome. People who are both strongly risk averse and future-biased can face a series of choices that will guarantee them more pain, for no compensating benefit: (...) they will be pain pumped. Thus, combining these preferences is rationally impermissible. Dougherty (2011) argues that this gives us reason to think that future-bias is rationally impermissible. This argument, and a similar one presented by Greene and Sullivan (2015), relies on the idea that if several preferences are not rationally combinable, then in the absence of some explanation of why they are not combinable, we should conclude that (at least) one of them is rationally impermissible. We take up this question by, inter alia, reflecting on the empirical results of a study we ran that probes people’s risk averse and future-biased preferences. We argue that the data, in conjunction with other considerations, suggests that we should not infer from the irrationality of the combination of these preferences to the irrationality of one of the preferences singly. We also argue that given the descriptive data about these preferences, there is no reason to think that, if one of them is rationally impermissible, then it is future-bias. (shrink)

This article develops a constructive criticism of methodological conventionalism. Methodological conventionalism asserts that standards of inductive risk ought to be justified in virtue of their ability to facilitate coordination in a research community. On that view, industry bias occurs when conventional methodological standards are violated to foster industry preferences. The underlying account of scientific conventionality, however, is problematically incomplete. Conventions may be justified in virtue of their coordinative functions, but often qualify for posterior empirical criticism as research advances. (...) Accordingly, industry bias does not only threaten existing conventions but may impede their empirically warranted improvement if they align with industry preferences. My empiricist account of standards of inductive risk avoids such a problem by asserting that conventional justification can be pragmatically warranted but has, in principle, only a provisional status. Methodological conventions, therefore, should not only be defended from preference-based infringements on their coordinative function but ought to be subjected to empirical criticism. (shrink)

Risk Management Strategies and Investment Behaviors are considered important factors in the investing activities of the retail investors. This study seeks to determine the relationship between Risk Management Strategies and Investment Behavior of Generation Z retail investors. The study is a correlational research and purposive sampling was used to select the respondents for this study. Cochran’s formula was utilized to determine the total sample size or total number of respondents. Spearman’s Rank-Order Correlation was employed to assess the significant (...) relationship of Risk Management Strategies and Investment Behaviors. Based on the results, it showed that risk management strategies have a significant relationship with one or more of the investment behaviors. Risk avoidance has a significant relationship with confirmation bias. Risk reduction has a significant relationship with all of the investment behaviors such as overconfidence, loss-aversion, recency bias, and confirmation bias. Risk transfer shows a significant relationship with only loss-aversion and confirmation bias. And lastly, risk acceptance also has a significant relationship with all of the investment behaviors. This study recommends exploring other types of behavioral biases under cognitive bias and emotional bias to determine their correlation with risk management strategies. Also, broadening the sample size, demographic scope, and geographic scope of the study can improve the generalizability of the study and provide a more comprehensive understanding of the relationship between investment behavior and risk management strategies. The researchers also recommend to utilize interviews and focus group discussions to gain an in-depth understanding of the underlying values and attitudes concerning investment behavior. (shrink)

Special Issue “Risks of artificial general intelligence”, Journal of Experimental and Theoretical Artificial Intelligence, 26/3 (2014), ed. Vincent C. Müller. http://www.tandfonline.com/toc/teta20/26/3# - Risks of general artificial intelligence, Vincent C. Müller, pages 297-301 - Autonomous technology and the greater human good - Steve Omohundro - pages 303-315 - - - The errors, insights and lessons of famous AI predictions – and what they mean for the future - Stuart Armstrong, Kaj Sotala & Seán S. Ó hÉigeartaigh - pages 317-342 - - (...) - The path to more general artificial intelligence - Ted Goertzel - pages 343-354 - - - Limitations and risks of machine ethics - Miles Brundage - pages 355-372 - - - Utility function security in artificially intelligent agents - Roman V. Yampolskiy - pages 373-389 - - - GOLEM: towards an AGI meta-architecture enabling both goal preservation and radical self-improvement - Ben Goertzel - pages 391-403 - - - Universal empathy and ethical bias for artificial general intelligence - Alexey Potapov & Sergey Rodionov - pages 405-416 - - - Bounding the impact of AGI - András Kornai - pages 417-438 - - - Ethics of brain emulations - Anders Sandberg - pages 439-457. (shrink)

Papers from the conference on AI Risk (published in JETAI), supplemented by additional work. --- If the intelligence of artificial systems were to surpass that of humans, humanity would face significant risks. The time has come to consider these issues, and this consideration must include progress in artificial intelligence (AI) as much as insights from AI theory. -- Featuring contributions from leading experts and thinkers in artificial intelligence, Risks of Artificial Intelligence is the first volume of collected chapters dedicated (...) to examining the risks of AI. The book evaluates predictions of the future of AI, proposes ways to ensure that AI systems will be beneficial to humans, and then critically evaluates such proposals. 1 Vincent C. Müller, Editorial: Risks of Artificial Intelligence - 2 Steve Omohundro, Autonomous Technology and the Greater Human Good - 3 Stuart Armstrong, Kaj Sotala and Sean O’Heigeartaigh, The Errors, Insights and Lessons of Famous AI Predictions - and What they Mean for the Future - 4 Ted Goertzel, The Path to More General Artificial Intelligence - 5 Miles Brundage, Limitations and Risks of Machine Ethics - 6 Roman Yampolskiy, Utility Function Security in Artificially Intelligent Agents - 7 Ben Goertzel, GOLEM: Toward an AGI Meta-Architecture Enabling Both Goal Preservation and Radical Self-Improvement - 8 Alexey Potapov and Sergey Rodionov, Universal Empathy and Ethical Bias for Artificial General Intelligence - 9 András Kornai, Bounding the Impact of AGI - 10 Anders Sandberg, Ethics and Impact of Brain Emulations 11 Daniel Dewey, Long-Term Strategies for Ending Existential Risk from Fast Takeoff - 12 Mark Bishop, The Singularity, or How I Learned to Stop Worrying and Love AI -. (shrink)

Some experts have argued that patients facing certain types of choices should not be told whether their risk is above or below average, because this information may trigger a bias (Fagerlin et al. 2007). But careful consideration shows that the comparative risk heuristic can usefully guide decisions and improve their quality or rationality. Building on an earlier paper of mine (Schwartz 2009), I will argue here that doctors and decision aids should provide comparative risk information to (...) patients, even while further research is conducted. (shrink)

As algorithms have become ubiquitous in consequential domains, societal concerns about the potential for discriminatory outcomes have prompted urgent calls to address algorithmic bias. In response, a rich literature across computer science, law, and ethics is rapidly proliferating to advance approaches to designing fair algorithms. Yet computer scientists, legal scholars, and ethicists are often not speaking the same language when using the term ‘bias.’ Debates concerning whether society can or should tackle the problem of algorithmic bias are (...) hampered by conflations of various understandings of bias, ranging from neutral deviations from a standard to morally problematic instances of injustice due to prejudice, discrimination, and disparate treatment. This terminological confusion impedes efforts to address clear cases of discrimination. -/- In this paper, we examine the promises and challenges of different approaches to disambiguating bias and designing for justice. While both approaches aid in understanding and addressing clear algorithmic harms, we argue that they also risk being leveraged in ways that ultimately deflect accountability from those building and deploying these systems. Applying this analysis to recent examples of generative AI, our argument highlights unseen dangers in current methods of evaluating algorithmic bias and points to ways to redirect approaches to addressing bias in generative AI at its early stages in ways that can more robustly meet the demands of justice. (shrink)

Some artificial intelligence systems can display algorithmic bias, i.e. they may produce outputs that unfairly discriminate against people based on their social identity. Much research on this topic focuses on algorithmic bias that disadvantages people based on their gender or racial identity. The related ethical problems are significant and well known. Algorithmic bias against other aspects of people’s social identity, for instance, their political orientation, remains largely unexplored. This paper argues that algorithmic bias against people’s political (...) orientation can arise in some of the same ways in which algorithmic gender and racial biases emerge. However, it differs importantly from them because there are strong social norms against gender and racial biases. This does not hold to the same extent for political biases. Political biases can thus more powerfully influence people, which increases the chances that these biases become embedded in algorithms and makes algorithmic political biases harder to detect and eradicate than gender and racial biases even though they all can produce similar harm. Since some algorithms can now also easily identify people’s political orientations against their will, these problems are exacerbated. Algorithmic political bias thus raises substantial and distinctive risks that the AI community should be aware of and examine. (shrink)

The reasonable person standard is used in adjudicating claims of self-defence. In US law, an individual may use defensive force if her beliefs that a threat is imminent and that force is required are beliefs that a reasonable person would have. In English law, it is sufficient that beliefs in imminence and necessity are genuinely held; but the reasonableness of so believing is given an evidential role in establishing the genuineness of the beliefs. There is, of course, much contention over (...) how to spell out when, and in virtue of what, such beliefs are reasonable. In this chapter, we identify some distinctive issues that arise when we consider that implicit racial bias might be implicated in the beliefs in imminence and necessity. Considering two prominent interpretations of the reasonable person standard, we argue that neither is acceptable. On one interpretation, we risk unfairness to the defendant-who may non-culpably harbour bias. On another, the standard embeds racist stereotypes. Whilst there are formulations of the defence that may serve to mitigate these problems, we argue that they cannot be avoided in the presence of racist social structures. (shrink)

Many philosophers think that we should use a lottery to decide who gets a good to which two persons have an equal claim but which only one person can get. Some philosophers think that we should save identified persons from harm even at the expense of saving a somewhat greater number of statistical persons from the same harm. I defend a principled way of justifying both judgements, namely, by appealing to the decreasing marginal moral value of survival chances. I identify (...) four desiderata that, I contend, any such justification should satisfy, and explain how my account meets these desiderata, unlike some previous accounts. (shrink)

Conservative assumptions in medical ethics risk immense harms during a pandemic. Public health institutions and public discourse alike have repeatedly privileged inaction over aggressive medical interventions to address the pandemic, perversely increasing population-wide risks while claiming to be guided by ‘caution’. This puzzling disconnect between rhetoric and reality is suggestive of an underlying philosophical confusion. In this paper, I argue that we have been misled by status quo bias—exaggerating the moral significance of the risks inherent in medical interventions, (...) while systematically neglecting the risks inherent in the status quo prospect of an out-of-control pandemic. By coming to appreciate the possibility and significance of status quo risk, we will be better prepared to respond appropriately when the next pandemic strikes. (shrink)

The chapter discusses two points of intersection between the communication of science-based information about risk and philosophical ethics. The first is a logically unnecessary bias toward consequentialist ethics, and a corresponding tendency to overlook the significance of deontological and virtue based ways to interpret the findings of a scientific risk analysis. The second is a grammatical bias that puts scientific communicators at odds with the expectations of a non-scientific audience.

Outbreaks of novel influenza viruses are continually occurring on many places on our planet, with the ultimate and most extreme consequence being a full-scale pandemic. Modern communication technology is widely used for risk communication regarding recommended change in behavior patterns and other precautions in order to mitigate the transmission. However, the assumption and bias that modern communication technology constitutes the norm causes vulnerable groups to be at possible risk of systematic exclusion to correct and updated information. Through (...) conducting a literature- and case analysis, the aim of this study is to identify insufficient or inadequate risk communication efforts in South Korea and Vietnam during influenza outbreaks, especially with concern of vulnerable groups. Further, to analyze how national influenza preparedness plans observe or ignore these insufficiencies. Results show that vulnerable groups are explicitly recognized in the preparedness plan of Vietnam. However, the South Korean preparedness plan show a more homogenous approach. Both South Korea and Vietnam showed a broad variety of channels used in their risk communication strategies which could be positive in terms of a broad outreach to a heterogenous population, including vulnerable groups. Four key factors that moderate the outcomes of risk communication were identified: Channels, Messages, Transparency and Trust. (shrink)

This literature review explores the issues and risks in non-conscious data collection and evaluates people’s attitudes towards it. In the modern world, data is one of the most valuable resources, yet studies focused on the potential negative implications of the new data-driven technologies are lacking. Therefore, this thesis conducts a comprehensive literature review to identify and assess risks in non-conscious data collection technologies that are most relevant and referenced in current literature. Accordingly, the most prominent risks are related to privacy (...) issues with personal data, bias in algorithms creating inequality, and the difficulties in creating adequate legislation. Subsequently, the thesis will explore existing studies about people’s attitudes towards non-conscious data collection and examine the most significant socio-demographic determinants of those attitudes. Therefore, despite socio-demographic factors such as age, gender, and economic status affecting the attitudes people have towards data collection technologies, this thesis argues how these approaches can be explained by varying external factors such as prior experiences with similar technologies. (shrink)

English abstract: Feminist philosophers of science have argued that various biases can and do influence the results of scientific investigations. Two kinds of arguments have been most influential: On the one hand, it has been argued that biased assumptions frequently bridge the gap between observation and theory associated with ‘the underdetermination thesis’. On the other hand, it has been argued that biased value judgments determine when the evidence in favor of a particular theory is considered sufficiently strong for the theory (...) to be accepted as true. This paper argues that bias can influence the results of scientific investigations via a third route. Briefly, biases can and do influence which theories are seriously proposed and considered in scientific communities, which in turn leads to otherwise promising theories being ignored when scientists are choosing between such theories. In the final section, I suggest that this result speaks in favor of taking a quite radical approach to eliminating unwanted biases in science. (shrink)

Measures of inductive risk and of safety-principle violation help us to operationalize concerns about theological assertions or a sort which, as we saw in Part I, aggravate or intensify problems of religious luck. Our overall focus in Part II will remain on a) responses to religious multiplicity, and b) sharply asymmetrical religious trait-ascriptions to religious insiders and outsiders. But in Part II formal markers of inductive norm violation will supply an empirically-based manner of distinguishing strong from moderate fideism. As (...) we develop these markers we will elaborate their more specific connections with comparative study of religious fundamentalisms (chapters 3 and 4), with exclusivist responses to religious multiplicity (chapter 5), and with working hypotheses in cognitive science of religion (chapter 6). In Chapters 3 the special focus is on the need for comparative fundamentalism (hereafter CF), and on how a better inductive risk ‘toolkit’ can empower its development. The “Enemy in the Mirror” is a metaphor which researchers of CF have sometimes used to describe a phenomena of special concern. This allows that religious fundamentalism per se need not be morally or socially problem, and that the terms such as “fundamentalism” and “fideism should not be over-used by scholars. But the enemy-in-the-mirror phenomena, which gives rise to what I term “bias-mirroring” attributions of good/bad traits to religious insiders and outsiders, carries enormous moral risks. I argue that this is something researchers would do well to study. On the view to be developed, the enemy in the mirror phenomenon is a direct consequent of counter-inductive thinking when applied to a multiplicity of narrative testimonial traditions. (shrink)

Artificial intelligence (AI) is a digital technology that will be of major importance for the development of humanity in the near future. AI has raised fundamental questions about what we should do with such systems, what the systems themselves should do, what risks they involve and how we can control these. - After the background to the field (1), this article introduces the main debates (2), first on ethical issues that arise with AI systems as objects, i.e. tools made and (...) used by humans; here, the main sections are privacy (2.1), manipulation (2.2), opacity (2.3), bias (2.4), autonomy & responsibility (2.6) and the singularity (2.7). Then we look at AI systems as subjects, i.e. when ethics is for the AI systems themselves in machine ethics (2.8.) and artificial moral agency (2.9). Finally we look at future developments and the concept of AI (3). For each section within these themes, we provide a general explanation of the ethical issues, we outline existing positions and arguments, then we analyse how this plays out with current technologies and finally what policy conse-quences may be drawn. (shrink)

The main aim of this article is to reflect on the impact of biases related to artificial intelligence (AI) systems developed to tackle issues arising from the COVID-19 pandemic, with special focus on those developed for triage and risk prediction. A secondary aim is to review assessment tools that have been developed to prevent biases in AI systems. In addition, we provide a conceptual clarification for some terms related to biases in this particular context. We focus mainly on nonracial (...) biases that may be less considered when addressing biases in AI systems in the existing literature. In the manuscript, we found that the existence of bias in AI systems used for COVID-19 can result in algorithmic justice and that the legal frameworks and strategies developed to prevent the apparition of bias have failed to adequately consider social determinants of health. Finally, we make some recommendations on how to include more diverse professional profiles in order to develop AI systems that increase the epistemic diversity needed to tackle AI biases during the COVID-19 pandemic and beyond. (shrink)

In many assessment problems—aptitude testing, hiring decisions, appraisals of the risk of recidivism, evaluation of the credibility of testimonial sources, and so on—the fair treatment of different groups of individuals is an important goal. But individuals can be legitimately grouped in many different ways. Using a framework and fairness constraints explored in research on algorithmic fairness, I show that eliminating certain forms of bias across groups for one way of classifying individuals can make it impossible to eliminate such (...) bias across groups for another way of dividing people up. And this point generalizes if we require merely that assessments be approximately bias-free. Moreover, even if the fairness constraints are satisfied for some given partitions of the population, the constraints can fail for the coarsest common refinement, that is, the partition generated by taking intersections of the elements of these coarser partitions. This shows that these prominent fairness constraints admit the possibility of forms of intersectional bias. (shrink)

Those who care about and engage in politics frequently fall victim to cognitive bias. Concerns that such bias impacts scholarship recently have prompted debates—notably, in philosophy and psychology—on the proper relationship between research and politics. One proposal emerging from these debates is that researchers studying politics have a professional duty to avoid political activism because it risks biasing their work. While sympathetic to the motivations behind this proposal, I suggest several reasons to reject a blanket duty to avoid (...) activism: (1) even if it reduced bias, this duty would make unreasonable demands on researchers; (2) this duty could hinder research by limiting viewpoint diversity; (3) this duty wrongly implies that academia offers a relative haven from bias compared to politics; and (4) not all forms of political activism pose an equal risk of bias. None of these points suggest that researchers should ignore the risk of bias. Rather, researchers should focus on stronger evidence-based strategies for reducing bias than a blanket recommendation to avoid politics. (shrink)

We often show a greater inclination to assist and avoid harming people identified as those at high risk of great harm than to assist and avoid harming people who will suffer similar harm but are not identified (as yet). Call this the identified person bias. Some ethicists think such bias is justified; others disagree and claim that the bias is discriminatory against statistical people. While the issue is present in public policy and politics, perhaps the most (...) notable examples can be found in medical ethics such as in ICU triage decisions made during the COVID‐19 pandemic. The Rule of Rescue, as the application of the identified person bias is sometimes called, states that spending large amounts of resources rescuing identifiable individuals who are in imminent danger is justified. In this paper, I show that our distorted attitudes toward time play a role in identified person bias. I claim that ICU triage decisions are better explained by a preference to treat people sooner rather than later, which is at least partially due to near bias (positive events are to be preferred to be near rather than distant), than by a preference to treat identified lives over statistical lives. Thus, another bias, near bias, is involved in the reasoning behind the identified person bias and the Rule of Rescue. (shrink)

This book has argued that problems of religious luck, especially when operationalized into concerns about doxastic risk and responsibility, can be of shared interest to theologians, philosophers, and psychologists. We have pointed out counter-inductive thinking as a key feature of fideistic models of faith, and examined the implications of this point both for the social scientific study of fundamentalism, and for philosophers’ and theologians’ normative concerns with the reasonableness of a) exclusivist attitudes to religious multiplicity, and b) theologically-cast but (...) bias-mirroring trait-ascriptions to religious insiders and outsiders. It is important to keep the descriptive/explanatory and normative concerns properly separated, but philosophy of luck and risk are relevant to both. More specifically, inductive risky theological strategies,we have argued, are a relevant concern both descriptively and normatively. The descriptive/explanatory relevance of measures of high inductive risk connects it with cognitive and social psychology of religion, while its normative relevance connects with critical concerns with epistemology of testimony, the epistemology of disagreement, and the ethics of belief. A research program to examine fideistic orientation and its relation to epistemically risky doxastic strategies is one of potentially numerous research programs on which philosophers and psychologists might work collaboratively. So this concluding chapter of our study culminates with the outline of a proposed research program at the intersection of shared concerns. I term this research program CICI, because it examines what lies at the intersection of CSR’s standing interest in the appeal of counter-intuitive ideas, and our own study’s focus on the fideistic penchant for counter-inductive thinking. Religious Studies scholars typically focus on particular traditions and teachings, while CSR scholars tend to eschew such content-focused approaches in favor of a study of evolutionary and hence generic or trans-religious functions and processes. I argue that CICI has the added benefit of effectively mediating this generic-specific contrast between CSR and Religious Studies, allowing CSR research to be more closely connected with and relevant to comparative fundamentalism. (shrink)

In its recent statement 'Sex Selection and Preimplantation Genetic Diagnosis', the Ethics Committee of the American Society of Reproductive Medicine concluded that preimplantation genetic diagnosis for sex selection for non-medical reasons should be discouraged because it poses a risk of unwarranted gender bias, social harm, and results in the diversion of medical resources from genuine medical need. We critically examine the arguments presented against sex selection using preimplantation genetic diagnosis. We argue that sex selection should be available, at (...) least within privately funded health care. (shrink)

Artificial intelligence (AI) and robotics are digital technologies that will have significant impact on the development of humanity in the near future. They have raised fundamental questions about what we should do with these systems, what the systems themselves should do, what risks they involve, and how we can control these. - After the Introduction to the field (§1), the main themes (§2) of this article are: Ethical issues that arise with AI systems as objects, i.e., tools made and used (...) by humans. This includes issues of privacy (§2.1) and manipulation (§2.2), opacity (§2.3) and bias (§2.4), human-robot interaction (§2.5), employment (§2.6), and the effects of autonomy (§2.7). Then AI systems as subjects, i.e., ethics for the AI systems themselves in machine ethics (§2.8) and artificial moral agency (§2.9). Finally, the problem of a possible future AI superintelligence leading to a “singularity” (§2.10). We close with a remark on the vision of AI (§3). - For each section within these themes, we provide a general explanation of the ethical issues, outline existing positions and arguments, then analyse how these play out with current technologies and finally, what policy consequences may be drawn. (shrink)

Recent existential-risk thinkers have noted that the analysis of the fine-tuning argument for God’s existence, and the analysis of certain forms of existential risk, employ similar types of reasoning. This paper argues that insofar as the “many worlds objection” undermines the inference to God’s existence from universal fine-tuning, then a similar many worlds objection undermines the inference that the historic risk of global nuclear catastrophe has been low from the lack of such a catastrophe has occurred in (...) our world. A version of the fine-tuning argument applied to nuclear risk, The Nuclear Fine-Tuning Argument, utilizes the set of nuclear close calls to show that 1) conventional explanations fail to adequately explain how we have survived thus far and 2) the existence of many worlds provides an adequate explanation. This is because, if there are many worlds, observers are disproportionately more likely to reflect upon a world that hasn’t had a global nuclear catastrophe than upon one that has had a global nuclear catastrophe. This selection bias results from the catastrophic nature of such an event. This argument extends generally to all global catastrophic risks that both A) have been historic threats and B) would result in a significantly lower global population. (shrink)

Mental accounting is a concept associated with the work of Richard Thaler. According to Thaler, people think of value in relative rather than absolute terms. They derive pleasure not just from an object’s value, but also the quality of the deal – its transaction utility (Thaler, 1985). In addition, humans often fail to fully consider opportunity costs (tradeoffs) and are susceptible to the sunk cost fallacy. Why are people willing to spend more when they pay with a credit card than (...) cash (Prelec & Simester, 2001)? Why would more individuals spend $10 on a theater ticket if they had just lost a $10 bill than if they had to replace a lost ticket worth $10 (Kahneman & Tversky, 1984)? Why are people more likely to spend a small inheritance and invest a large one (Thaler, 1985)? According to the theory of mental accounting, people treat money differently, depending on factors such as the money’s origin and intended use, rather than thinking of it in terms of the “bottom line” as in formal accounting (Thaler, 1999). An important term underlying the theory is fungibility, the fact that all money is interchangable and has no labels. In mental accounting, people treat assets as less fungible than they really are. Even seasoned investors are susceptible to this bias when they view recent gains as disposable “house money” (Thaler & Johnson, 1990) that can be used in high-risk investments. In doing so, they make decisions on each mental account separately, losing out the big picture of the portfolio. (See also partitioning and pain of paying for ideas related to mental accounting.) . (shrink)

Most of us are “time-biased” in preferring pains to be past rather than future and pleasures to be future rather than past. However, it turns out that if you are risk averse and time-biased, then you can be turned into a “pain pump”—in order to insure yourself against misfortune, you will take a series of pills which leaves you with more pain and better off in no respect. Since this vulnerability seems rationally impermissible, while time-bias and risk (...) aversion seem rationally permissible, we are left with a puzzle. (shrink)

When agents insert technological systems into their decision-making processes, they can obscure moral responsibility for the results. This can give rise to a distinct moral wrong, which we call “agency laundering.” At root, agency laundering involves obfuscating one’s moral responsibility by enlisting a technology or process to take some action and letting it forestall others from demanding an account for bad outcomes that result. We argue that the concept of agency laundering helps in understanding important moral problems in a number (...) of recent cases involving automated, or algorithmic, decision-systems. We apply our conception of agency laundering to a series of examples, including Facebook’s automated advertising suggestions, Uber’s driver interfaces, algorithmic evaluation of K-12 teachers, and risk assessment in criminal sentencing. We distinguish agency laundering from several other critiques of information technology, including the so-called “responsibility gap,” “bias laundering,” and masking. (shrink)

In this peer commentary on L. Syd M. Johnson’s “Inference and Inductive Risk in Disorders of Consciousness,” I argue for the necessity of disability education as an integral component of decision-making processes concerning patients with DOC and, mutatis mutandis, all patients with disabilities. The sole qualification Johnson places on such decision-making is that stakeholders are educated about and “understand the uncertainties of diagnosis and prognosis.” Drawing upon research in philosophy of disability, social epistemology, and health psychology, I argue that (...) this educational qualification is insufficient to address systemic ableism and other forms of epistemic bias in quality of life judgments. (shrink)

Over the years, companies have adopted hiring algorithms because they promise wider job candidate pools, lower recruitment costs and less human bias. Despite these promises, they also bring perils. Using them can inflict unintentional harms on individual human rights. These include the five human rights to work, equality and nondiscrimination, privacy, free expression and free association. Despite the human rights harms of hiring algorithms, the AI ethics literature has predominantly focused on abstract ethical principles. This is problematic for two (...) reasons. First, AI principles have been criticized for being vague and not actionable. Second, the use of vague ethical principles to discuss algorithmic risks does not provide any accountability. This lack of accountability creates an algorithmic accountability gap. Closing this gap is crucial because, without accountability, the use of hiring algorithms can lead to discrimination and unequal access to employment opportunities. This paper makes two contributions to the AI ethics literature. First, it frames the ethical risks of hiring algorithms using international human rights law as a universal standard for determining algorithmic accountability. Second, it evaluates four types of algorithmic impact assessments in terms of how effectively they address the five human rights of job applicants implicated in hiring algorithms. It determines which of the assessments can help companies audit their hiring algorithms and close the algorithmic accountability gap. (shrink)

This paper critically examines the philosophical underpinnings of current experimental investigation into animal affect-related decision-making. Animals’ affective states are standardly operationalised by linking positively valenced states with “approach” behaviours and negatively valenced states with “avoidance” behaviours. While this operationalisation has provided a helpful starting point to investigate the ecological role of animals’ internal states, there is extensive evidence that valenced and motivational states do not always neatly align, namely, instances where “liking” does not entail “wanting” (and vice versa). To address (...) this limitation, this paper proposes the (re-)integration of dominance as a dimension of affect. In particular, I argue that dominance, construed as a measure of organisms’ self-assessed behavioural control, can provide the necessary nuance to understand animal affect-related decision-making. Thus, after addressing the theoretical and methodological challenges related to incorporating dominance into a definition of “core affect”, this paper introduces the “Valence-Arousal-Dominance” model of animal affect. This model is explored for its potential applications in two domains. Firstly, in the study of animal affect-related decision-making under predation risk, and secondly, in the study of animal wellbeing. Through these applications, this model aims to bring experimental paradigms of animal affect-related behaviour closer to ecologically relevant scenarios. (shrink)

It is often thought that traditional recidivism prediction tools used in criminal sentencing, though biased in many ways, can straightforwardly avoid one particularly pernicious type of bias: direct racial discrimination. They can avoid this by excluding race from the list of variables employed to predict recidivism. A similar approach could be taken to the design of newer, machine learning-based (ML) tools for predicting recidivism: information about race could be withheld from the ML tool during its training phase, ensuring that (...) the resulting predictive model does not use race as an explicit predictor. However, if race is correlated with measured recidivism in the training data, the ML tool may ‘learn’ a perfect proxy for race. If such a proxy is found, the exclusion of race would do nothing to weaken the correlation between risk (mis)classifications and race. Is this a problem? We argue that, on some explanations of the wrongness of discrimination, it is. On these explanations, the use of an ML tool that perfectly proxies race would (likely) be more wrong than the use of a traditional tool that imperfectly proxies race. Indeed, on some views, use of a perfect proxy for race is plausibly as wrong as explicit racial profiling. We end by drawing out four implications of our arguments. (shrink)

A common response to claims of bias, harassment, or discrimination is to say that these claims are made in bad faith. Claimants are supposedly not motivated by a credible or even sincere belief that unfair or unequal treatment has occurred, but simply seek to illicitly gain public sympathy or private reward. Characterizing discrimination claims as systematically made in bad faith enables them to be screened and dismissed prior to engaging with them on their merits. This retort preserves the dominant (...) group’s self-image as unprejudiced and innocent without having to risk critical analysis of the claim’s substance. (shrink)

A self-fulfilling prophecy in neuroprognostication occurs when a patient in coma is predicted to have a poor outcome, and life-sustaining treatment is withdrawn on the basis of that prediction, thus directly bringing about a poor outcome for that patient. In contrast to the predominant emphasis in the bioethics literature, we look beyond the moral issues raised by the possibility that an erroneous prediction might lead to the death of a patient who otherwise would have lived. Instead, we focus on the (...) problematic epistemic consequences of neuroprognostic SFPs in settings where research and practice intersect. When this sort of SFP occurs, the problem is that physicians and researchers are never in a position to notice whether their original prognosis was correct or incorrect, since the patient dies anyway. Thus, SFPs keep us from discerning false positives from true positives, inhibiting proper assessment of novel prognostic tests. This epistemic problem of SFPs thus impedes learning, but ethical obligations of patient care make it difficult to avoid SFPs. We then show how the impediment to catching false positive indicators of poor outcome distorts research on novel techniques for neuroprognostication, allowing biases to persist in prognostic tests. We finally highlight a particular risk that a precautionary bias towards early withdrawal of life-sustaining treatment may be amplified. We conclude with guidelines about how researchers can mitigate the epistemic problems of SFPs, to achieve more responsible innovation of neuroprognostication for patients in coma. (shrink)

Proponents of epistemic partiality in friendship argue that friendship makes demands of our epistemic lives that are at least inconsistent with the demands of epistemic propriety, and perhaps downright irrational. In this paper, I focus on the possibility that our commitments to our friends distort how we respond to testimony about them, their character, and their conduct. Sometimes friendship might require us to ignore (or substantially underweight) what others tell us about our friends. However, while this practice might help promote (...) your friendship and protect your friends, refusing someone’s testimony can constitute a special kind of insult towards them, according to Allen Hazlett (2017) and Finlay Malcolm (2018). It seems to follow that friendship can require us to insult people by refusing their testimony. This should count against treating your friends with epistemic partiality, one might think. It’s plausible that we have moral reasons not to insult other people, whether they’re our friends or not. The risk of insulting non-friends, it turns out, is just the beginning of the problems for the epistemic partiality in friendship. Recent discussions in social epistemology, led by Miranda Fricker (2007), Kristie Dotson (2011) and José Medina (2012), among others, have brought renewed attention the distinctly epistemic nature of some kinds of injustice. Drawing on this work, I’ll suggest a principled way to determine when epistemic partiality in friendship is morally impermissible because of the genuine risk of harmful epistemic injustice. -/- The paper proceeds as follows. First, I’ll argue that treating our friends with epistemic partiality can involve insulting other people by refusing their testimony. Then I’ll consider whether we ever ought to commit more morally serious testimonial injustices in the name of friendship. After considering some putative cases, I’ll introduce the concept of testimonial injustice, referring primarily to the work of Dotson (2011). I’ll explain how and why the epistemic bias which I’ve prescribed in friendship could be thought to cause such injustice. Drawing on the work of Sheila Lintott (2015), I’ll consider some ways in which the extent of the testimonial injustice caused by bias in friendship might be worse yet, namely, because of the role that homophily plays in determining who we become (and remain) friends with. In particular, I’ll consider whether epistemic partiality perpetuates racist or sexist forms of testimonial injustice. Finally, I’ll attempt to alleviate some of these concerns and determine when we’re morally required to treat friends with epistemic partiality. I’ll argue that whilst protecting a friend’s well-being might require you to cause testimonial injustice, such injustice would be morally unjustified. Accordingly, if you commit such injustice in the name of friendship, you’re morally blameworthy for doing so. This, I suggest, imposes an additional limit on epistemic partiality in friendship. (shrink)

What does Artificial Intelligence (AI) have to contribute to health care? And what should we be looking out for if we are worried about its risks? In this paper we offer a survey, and initial evaluation, of hopes and fears about the applications of artificial intelligence in medicine. AI clearly has enormous potential as a research tool, in genomics and public health especially, as well as a diagnostic aid. It’s also highly likely to impact on the organisational and business practices (...) of healthcare systems in ways that are perhaps under-appreciated. Enthusiasts for AI have held out the prospect that it will free physicians up to spend more time attending to what really matters to them and their patients. We will argue that this claim depends upon implausible assumptions about the institutional and economic imperatives operating in contemporary healthcare settings. We will also highlight important concerns about privacy, surveillance, and bias in big data, as well as the risks of over trust in machines, the challenges of transparency, the deskilling of healthcare practitioners, the way AI reframes healthcare, and the implications of AI for the distribution of power in healthcare institutions. We will suggest that two questions, in particular, are deserving of further attention from philosophers and bioethicists. What does care look like when one is dealing with data as much as people? And, what weight should we give to the advice of machines in our own deliberations about medical decisions? (shrink)

For children, the collateral damage of the COVID-19 pandemic response has been considerable. In this paper, we use the framework of evidence-based medicine to argue that child abuse is another negative side effect of COVID-19 lockdowns. While it was certain that school closures would have profound social and economic costs, it remains uncertain whether they have any effect on COVID-19 transmission. There is emerging evidence that lockdowns significantly worsened child abuse on a global scale. Low-income and middle-income countries are particularly (...) vulnerable to increases in child abuse. The best available external evidence from systematic research during the pandemic demonstrates an increase in the risk of child maltreatment, an increase in child maltreatment hospitalisations and a concerning decrease in official child maltreatment referrals. The paradoxical phenomenon of increased hospitalisations and decreased reports is unlikely to be explained by a genuine decrease in child abuse. We conclude that lockdowns have an unacceptably high risk of negative side effects for children, as evidenced by child abuse, the true extent of which appears to be masked by lockdown-related disruptions to schools and other surveillance systems. The desire for a sense of security may be a tempting bias towards emphasising the resilience of children, but it is ethically problematic to push children towards abuse in the name of public health. It is our view that the collateral damage of prolonged school closures for society’s most vulnerable members is a powerful ethical consideration against any pandemic response which involves their use. (shrink)

One of the main lines of research in algorithmic fairness involves individual fairness (IF) methods. Individual fairness is motivated by an intuitive principle, similar treatment, which requires that similar individuals be treated similarly. IF offers a precise account of this principle using distance metrics to evaluate the similarity of individuals. Proponents of individual fairness have argued that it gives the correct definition of algorithmic fairness, and that it should therefore be preferred to other methods for determining fairness. I argue that (...) individual fairness cannot serve as a definition of fairness. Moreover, IF methods should not be given priority over other fairness methods, nor used in isolation from them. To support these conclusions, I describe four in-principle problems for individual fairness as a definition and as a method for ensuring fairness: (1) counterexamples show that similar treatment (and therefore IF) are insufficient to guarantee fairness; (2) IF methods for learning similarity metrics are at risk of encoding human implicit bias; (3) IF requires prior moral judgments, limiting its usefulness as a guide for fairness and undermining its claim to define fairness; and (4) the incommensurability of relevant moral values makes similarity metrics impossible for many tasks. In light of these limitations, I suggest that individual fairness cannot be a definition of fairness, and instead should be seen as one tool among several for ameliorating algorithmic bias. (shrink)

Recent studies have shown that recidivism scoring algorithms like COMPAS have significant racial bias: Black defendants are roughly twice as likely as white defendants to be mistakenly classified as medium- or high-risk. This has led some to call for abolishing COMPAS. But many others have argued that algorithms should instead be given access to a defendant's race, which, perhaps counterintuitively, is likely to improve outcomes. This approach can involve either establishing race-sensitive risk thresholds, or distinct racial ‘tracks’. (...) Is there a moral difference between these two approaches? We first consider Deborah Hellman's view that the use of distinct racial tracks (but not distinct thresholds) does not constitute disparate treatment since the effects on individuals are indirect and does not rely on a racial generalization. We argue that this is mistaken: the use of different racial tracks seems both to have direct effects on and to rely on a racial generalization. We then offer an alternative understanding of the distinction between these two approaches—namely, that the use of different cut points is to the counterfactual comparative disadvantage, ex ante, of all white defendants, while the use of different racial tracks can in principle be to the advantage of all groups, though some defendants in both groups will fare worse. Does this mean that the use of cut points is impermissible? Ultimately, we argue, while there are reasons to be skeptical of the use of distinct cut points, it is an open question whether these reasons suffice to make a difference to their moral permissibility. (shrink)

The philosophical literature on scientific explanation in neuroscience has been dominated by the idea of mechanisms. The mechanist philosophers often claim that neuroscience is in the business of finding mechanisms. This view has been challenged in numerous ways by showing that there are other successful and widespread explanatory strategies in neuroscience. However, the empirical evidence for all these claims was hitherto lacking. Empirical evidence about the pervasiveness and uses of various explanatory strategies in neuroscience is particularly needed because examples and (...) case studies that are used to illustrate philosophical claims so far tend to be hand-picked. The risk of confirmation bias is therefore considerable: when looking for white swans, all one finds is that swans are white. The more systematic quantitative and qualitative bibliometric study of a large body of relevant literature that we present in this paper can put such claims into perspective. Using bibliometric tools, we identify the typical linguistic patterns used in the alleged mechanistic, dynamical, and topological explanations in the literature, their preponderance and how they change over time. Our findings show abundant use of mechanistic language, but also the presence of a significant neuroscience literature using topological and dynamical explanatory language, which grows over time and increasingly differentiates from each other and from mechanistic explanations. (shrink)

This dissertation compares the notions of homo rationalis in Philosophy and homo oeconomicus in Economics. Particularly, in Part I, we claim that both notions are close methodological substitutes. Accordingly, we show that the constraints involved in the notion of economic rationality apply to the philosophical notion of rationality. On these premises, we explore the links between the notions of Kantian and Humean rationality in Philosophy and the constructivist and ecological approaches to rationality in economics, respectively. Particularly, we show that the (...) constraints involved in both approaches to rationality in economics apply to the notions of Kantian and Humean rationality. That is why we introduce the notion of minimal global rationality. In a nutshell, the latter links one’s rationality to one’s ability to make good use of their cognitive apparatus to process contextual variables. On this basis, in Part II, we present two empirical tests showing that minimal global rationality is at work in philosophical experiments because systematic manipulations of contextual variables predictably affect philosophical judgment in the Trolley Problem and in some Knobe-like vignettes. In the first case, the manipulation of the information provided to subjects predictably causes inconsistent judgments. More precisely, subjects’ response to information shocks predictably causes them to issue inconsistent judgments. Instead, in the second case, our experimental results reveal that the Knobe Effect activates only if the perpetrator of an action is likely to be perceived as evil and unfriendly. Besides, when the Knobe-like vignettes include probabilistic outcomes, the Knobe Effect does not activate or is reduced in magnitude. For these reasons, we maintain that the availability heuristic bias arguably affects one’s judgment in Knobe-like experiment settings. Eventually, in Part III, we survey the general implications of the findings presented in Part II. First and foremost, we show that when manned social media algorithms produce wide-ranging availability cascades, agents predictably respond to those cascades by herding around biased claims. Interestingly, we find that the exposure to similarly redundant online content makes new social groups emerge unhindered. That is why populists and conspiracy theorists obtain such a large follow-up online. Secondly, we find that manned social media algorithms can exploit the flaws of minimal global rationality at a faster pace than humans. Yet, given that latter algorithms cannot do so on their own, we argue that the Turing Test is still an open game. Specifically, we argue so because state-of-the-art technology can request a service to be supplied (e.g., Google Duplex) but cannot operate direct sales. This constraint of state-of-the-art technology is linked to its inability to exploit the flaws of global minimal rationality as well as humans can. Importantly, we show that if algorithms were able to do so, the involved sociopolitical risk would be high. Therefore, we argue that only some form of regulation over state-of-the-art technology can prevent and reduce such sociopolitical risks. -/- In case you want a copy of this dissertation, please email me. (shrink)

We propose a theoretical model to explain the usage of time-inconsistent behavior as a strategy to exploit others when reputation and trust have secondary effects on the economic outcome. We consider two agents with time-consistent preferences exploiting common resources. Supposing that an agent is believed to have time-inconsistent preferences with probability p, we analyze whether she uses this misinformation when she has the opportunity to use it. Using the model originally provided by Levhari and Mirman (Bell J Econ 11(1):322–334, 1980), (...) we determine the optimal degree of present bias that the agent would like to have while pretending to have time-inconsistent preferences and we provide the range of present-bias parameter under which deceiving is optimal. Moreover, by allowing the constant relative risk aversion class of utility form, we characterize the distinction between pretending to be naive and sophisticated. (shrink)

Introduction Current treatments for pain have limited benefits and worrying side effects. Some studies suggest that pain is reduced when clinicians deliver positive messages. However, the effects of positive messages are heterogeneous and have not been subject to meta-analysis. We aimed to estimate the efficacy of positive messages for pain reduction. -/- Methods We included randomized trials of the effects of positive messages in a subset of the studies included in a recent systematic review of context factors for treating pain. (...) Several electronic databases were searched. Reference lists of relevant studies were also searched. Two authors independently undertook study selection, data extraction, risk of bias assessment, and analyses. Our primary outcome measures were differences in patient- or observer-reported pain between groups who were given positive messages and those who were not. -/- Results Of the 16 randomized trials (1703 patients) that met the inclusion criteria, 12 trials had sufficient data for meta-analysis. The pooled standardized effect size was −0.31 (95% confidence interval [CI] −0.61 to −0.01, p = 0.04, I2 = 82%). The effect size remained positive but not statistically significant after we excluded studies considered to have a high risk of bias (standard effect size −0.17, 95% CI −0.54 to 0.19, P = 0.36, I2 = 84%). -/- Conclusion Care of patients with chronic or acute pain may be enhanced when clinicians deliver positive messages about possible clinical outcomes. However, we have identified several limitations of the present study that suggest caution when interpreting the results. We recommend further high-quality studies to confirm (or falsify) our result. -/- FUNDING -/- Alexander Mebius research has been funded through the ERC grant "Philosophy of Pharmacology: Safety, Statistical Standards, and Evidence Amalgamation" (GA: 639276). (shrink)

The central focus of this paper is to motivate and explore the question, when is it permissible to endorse a psychologizing explanation of a sincere interlocutor? I am interested in the moral question of when (if ever) we may permissibly dismiss the sincere reasons given to us by others, and instead endorse an alternative explanation of their beliefs and actions. I argue that there is a significant risk of wronging the other person, and so we should only psychologize when (...) we are in a position to know that they are in bad faith. (shrink)

In September 2020, Helsinki and Amsterdam announced the launch of their open AI registers—the first cities in the world to offer such a service. The AI registers describe what, where, and how AI applications are being used in the two municipalities; how algorithms were assessed for potential bias or risks; and how humans use the AI services. Examining issues of security and transparency, this paper discusses the potential for implementing AI in an urban public service setting and how this (...) might be harnessed to improve services and citizens’ experiences. -/- . (shrink)

Variations in people’s perceptions of investment risk and financial literacy have been linked in studies. More specifically, Diacon (2016) discovered significant differences between less financially savvy non-experts and financial professionals. Lay people therefore have a larger propensity for association bias (i.e., they give suppliers and salesmen a higher level of credibility than laypeople) and are often less risk-tolerant than financial professionals. The method of sampling that the researcher chose is known as purposeful sampling. According to Easton & (...) McColl, it is a fundamental sampling strategy where the researcher chooses a smaller group of people (a sample) from a larger group (a population) to study. The responders of the survey included both commissioned and non-commissioned PNP members. There were 67 non-commissioned individuals and 33 commissioned personnel. These individuals are at work while this study is being done. Because they think mutual funds can invest in a variety of assets in the future, think savings account interest is higher than that on fixed deposit accounts, and even budget their monthly income for expenses, savings, and investments, survey participants can be inferred to rate their level of financial knowledge as extremely high. Additionally, people learn about investments from their peers and think that growing costs will limit purchasing power. They also get banking advice from dependable friends. Because they want to make sure that they are not concerned about loans as long as their savings are protected and they refrain from purchasing non-essentials, police responders are persuaded that they have a very high attitude toward their finances. They also believe that investing in a business is a better use of their money than spending it now, that saving money over time is more satisfying than spending it now, and that their financial status is a big problem or a source of stress. The poll participants believe that their level of financial conduct is pretty high since the separate their needs and wants, have begun saving for their retirement, and consistently save money each pay period. They also budget their money, regulate their spending, and always pay their bills on time. (shrink)

The objective of this paper is to provide a psychological perspective on Zhu Xi (ZX) and Dai Zhen (DZ) views about human nature, by comparing the potential implications of their views on an agent's moral cultivation. To help frame this objective, I will ask and answer the following question: if one commits to ZX who holds the view that human nature is innately good, although obscured, versus if one holds DZ's view that while human nature has the potential for good (...) but it is unformed or unknown (i.e., no original nature) then what are some of the possible implications for self love, sympathy, hope, forgiveness, and spontaneity that are relevant considerations, some of which have been noted by ZX and DZ, for the advance of an agent's moral cultivation. The implications of ZX's commitment to human nature being innately good could entail the following: despite an agent’s obscurities, because his nature is good, he is lovable and he can be hopeful that he can shed off his obscurities via proper moral cultivation. Spontaneity is encouraged as an integral part of an agent's moral self-cultivation. His self-responsibility, hinges on his ability to use the instrumentality of moral cultivation, for which he would need the assistance of a moral teacher. There is a greater capacity for forgiveness because of the presumption that the human nature is inherently good. He can sympathize and extend concern for others, in part, because others' nature is also good. ZX's view may potentially carry a risk of excess and a risk of expecting mainly the good, but not the unknown. Alternatively, implications for DZ's commitment to no original human nature, entails the following: DZ's view is likely more conducive to expecting and embracing the unknown, which potentially makes DZ's philosophy more practical, because we live in a world where we often encounter unknowns and unfamiliar people. Self-love is a prerequisite to know love before one can love others. A moral agent can be hopeful because his potential is good, and it will not be a lost opportunity in light of the constitutive essence of moral cultivation. Despite DZ appearing to be against spontaneity, he is only against the kind of spontaneity that could be hurtful to others as does ZX. Lastly, I argue that DZ's view could result in a broader and more practical commitment to sympathy. Compared to ZX, I argue that DZ’s view could have a potential risk of lower self-responsibility and risk of resistance to self-forgiveness, which does not arise out of DZ’s views about the human nature per se, but rather stems from DZ's bias towards externalized morality. (shrink)