Ethics of Artificial Intelligence

Many researchers and intellectuals warn about extreme risks from artificial intelligence. However, these warnings typically came without systematic arguments in support. This paper provides an argument that AI will lead to the permanent disempowerment of humanity, e.g. human extinction, by 2100. It rests on four substantive premises which it motivates and defends: First, the speed of advances in AI capability, as well as the capability level current systems have already reached, suggest that it is practically possible to build AI systems (...) capable of disempowering humanity by 2100. Second, due to incentives and coordination problems, if it is possible to build such AI, it will be built. Third, since it appears to be a hard technical problem to build AI which is aligned with the goals of its designers, and many actors might build powerful AI, misaligned powerful AI will be built. Fourth, because disempowering humanity is useful for a large range of misaligned goals, such AI will try to disempower humanity. If AI is capable of disempowering humanity and tries to disempower humanity by 2100, then humanity will be disempowered by 2100. This conclusion has immense moral and prudential significance. (shrink)

The ongoing explosion of interest in artificial intelligence is fueled in part by recently developed techniques in machine learning. Those techniques allow automated systems to process huge amounts of data, utilizing mathematical methods that depart from traditional statistical approaches, and resulting in impressive advancements in our ability to make predictions and uncover correlations across a host of interesting domains. But as is now widely discussed, the way that those systems arrive at their outputs is often opaque, even to the experts (...) who design and deploy them. Is it morally problematic to make use of opaque automated methods when making high-stakes decisions, like whether to issue a loan to an applicant, or whether to approve a parole request? Many scholars answer in the affirmative. However, there is no widely accepted explanation for why transparent systems are morally preferable to opaque systems. We argue that the use of automated decision-making systems sometimes violates duties of consideration that are owed by decision-makers to decision-subjects, duties that are both epistemic and practical in character. Violations of that kind generate a weighty consideration against the use of opaque decision systems. In the course of defending our approach, we show that it is able to address three major challenges sometimes leveled against attempts to defend the moral import of transparency in automated decision-making. (shrink)

The prevailing discourse around AI ethics lacks the language and formalism necessary to capture the diverse ethical concerns that emerge when AI systems interact with individuals. Drawing on Sen and Nussbaum's capability approach, we present a framework formalizing a network of ethical concepts and entitlements necessary for AI systems to confer meaningful benefit or assistance to stakeholders. Such systems enhance stakeholders' ability to advance their life plans and well-being while upholding their fundamental rights. We characterize two necessary conditions for morally (...) permissible interactions between AI systems and those impacted by their functioning, and two sufficient conditions for realizing the ideal of meaningful benefit. We then contrast this ideal with several salient failure modes, namely, forms of social interactions that constitute unjustified paternalism, coercion, deception, exploitation and domination. The proliferation of incidents involving AI in high-stakes domains underscores the gravity of these issues and the imperative to take an ethics-led approach to AI systems from their inception. (shrink)

Fairness is central to the ethical and responsible development and use of AI systems, with a large number of frameworks and formal notions of algorithmic fairness being available. However, many of the fairness solutions proposed revolve around technical considerations and not the needs of and consequences for the most impacted communities. We therefore want to take the focus away from definitions and allow for the inclusion of societal and relational aspects to represent how the effects of AI systems impact and (...) are experienced by individuals and social groups. In this paper, we do this by means of proposing the ACROCPoLis framework to represent allocation processes with a modeling emphasis on fairness aspects. The framework provides a shared vocabulary in which the factors relevant to fairness assessments for different situations and procedures are made explicit, as well as their interrelations. This enables us to compare analogous situations, to highlight the differences in dissimilar situations, and to capture differing interpretations of the same situation by different stakeholders. (shrink)

Artificial intelligence (AI) has become indispensable in our lives. The development of a quantitative scale for AI ethics is necessary for a better understanding of public attitudes toward AI research ethics and to advance the discussion on using AI within society. For this study, we developed an AI ethics scale based on AI-specific scenarios. We investigated public attitudes toward AI ethics in Japan and the US using online questionnaires. We designed a test set using four dilemma scenarios and questionnaire items (...) based on a theoretical framework for ethics, legal, and social issues (ELSI). We found that _country_ and _age_ are the most informative sociodemographic categories for predicting attitudes for AI ethics. Our proposed scale, which consists of 13 questions, can be reduced to only three, covering ethics, tradition, and policies. This new AI ethics scale will help to quantify how AI research is accepted in society and which area of ELSI people are most concerned with. (shrink)

Artificial intelligence (AI) and machine learning (ML) systems increasingly purport to deliver knowledge about people and the world. Unfortunately, they also seem to frequently present results that repeat or magnify biased treatment of racial and other vulnerable minorities. This paper proposes that at least some of the problems with AI’s treatment of minorities can be captured by the concept of epistemic injustice. To substantiate this claim, I argue that (1) pretrial detention and physiognomic AI systems commit testimonial injustice because their (...) target variables reflect inaccurate and unjust proxies for what they claim to measure; (2) classification systems, such as facial recognition, commit hermeneutic injustice because their classification taxonomies, almost no matter how they are derived, reflect and perpetuate racial and other stereotypes; and (3) epistemic injustice better explains what is going wrong in these types of situations than does the more common focus on procedural (un)fairness. (shrink)

AI tools such as ChatGPT appear to magnify some of humanity’s worst qualities, and fixing those tendencies will be no easy task.

Machine learning algorithms are expected to improve referral decisions. In this article I discuss the legitimacy of deferring referral decisions in primary care to recommendations from such algorithms. The standard justification for introducing algorithmic decision procedures to make referral decisions is that they are more accurate than the available practitioners. The improvement in accuracy will ensure more efficient use of scarce health resources and improve patient care. In this article I introduce a proceduralist framework for discussing the legitimacy of algorithmic (...) referral decisions and I argue that in the context of referral decisions the legitimacy of an algorithmic decision procedure can be fully accounted for in terms of the instrumental values of accuracy and fairness. I end by considering how my discussion of procedural algorithmic legitimacy relates to the debate on algorithmic fairness. (shrink)

Algorithmic predictions are promising for insurance companies to develop personalized risk models for determining premiums. In this context, issues of fairness, discrimination, and social injustice might arise: Algorithms for estimating the risk based on personal data may be biased towards specific social groups, leading to systematic disadvantages for those groups. Personalized premiums may thus lead to discrimination and social injustice. It is well known from many application fields that such biases occur frequently and naturally when prediction models are applied to (...) people unless special efforts are made to avoid them. Insurance is no exception. In this paper, we provide a thorough analysis of algorithmic fairness in the case of insurance premiums. We ask what “fairness” might mean in this context and how the fairness of a premium system can be measured. For this, we apply the established fairness frameworks of the fair machine learning literature to the case of insurance premiums and show which of the existing fairness criteria can be applied to assess the fairness of insurance premiums. We argue that two of the often-discussed group fairness criteria, independence (also called statistical parity or demographic parity) and separation (also known as equalized odds), are not normatively appropriate for insurance premiums. Instead, we propose the sufficiency criterion (also known as well-calibration) as a morally defensible alternative that allows us to test for systematic biases in premiums towards certain groups based on the risk they bring to the pool. In addition, we clarify the connection between group fairness and different degrees of personalization. Our findings enable insurers to assess the fairness properties of their risk models, helping them avoid reputation damage resulting from potentially unfair and discriminatory premium systems. (shrink)

The unavailability of identification documents is a determining factor leading to social and economic exclusion for undocumented people. They cannot interact with public bodies and private subjects in an official way, so they cannot access services (healthcare, education, social welfare, etc.) or obtain formal employment. This sort of ‘identity gap’ between undocumented people and individuals with ID documents exacerbates socioeconomic discrepancies and inequalities and does not permit inclusive social development. Artificial intelligence represents a valid instrument in accomplishing the goal to (...) provide legal identity for all. AI algorithms could take care of several related tasks, such as identity authentication/validation, data matching and storage. However, using AI tools to collect and manage identity-related data comes with risks and drawbacks worth mentioning. Social and cultural influences contribute to the development of personal identity that is not limited to official documents. Thus, AI-driven technologies deployed in identity management should also consider such elements in performing their tasks. This chapter argues for adequate human rights safeguards when deploying AI algorithms to manage identity-related data. More specifically, this contribution calls for human oversight mechanisms and the periodical recalibration of such algorithms to address mutating environmental variables in the development of personal identity. (shrink)

This paper examines racial discrimination and algorithmic bias in predictive policing algorithms (PPAs), an emerging technology designed to predict threats and suggest solutions in law enforcement. We first describe what discrimination is in a case study of Chicago’s PPA. We then explain their causes with Broadbent’s contrastive model of causation and causal diagrams. Based on the cognitive science literature, we also explain why fairness is not an objective truth discoverable in laboratories but has context-sensitive social meanings that need to be (...) negotiated through democratic processes. With the above analysis, we next predict why some recommendations given in the bias reduction literature are not as effective as expected. Unlike the cliché highlighting equal participation for all stakeholders in predictive policing, we emphasize power structures to avoid hermeneutical lacunae. Finally, we aim to control PPA discrimination by proposing a governance solution—a framework of a social safety net. (shrink)

The present comment concerns a recent AI & Ethics article which purports to report evidence of speciesist bias in various popular computer vision (CV) and natural language processing (NLP) machine learning models described in the literature. I examine the authors' analysis and show it, ironically, to be prejudicial, often being founded on poorly conceived assumptions and suffering from fallacious and insufficiently rigorous reasoning, its superficial appeal in large part relying on the sequacity of the article's target readership.

Algorithm-assisted decision procedures—including some of the most high-profile ones, such as COMPAS—have been described as unfair because they compound injustice. The complaint is that in such procedures a decision disadvantaging members of a certain group is based on information reflecting the fact that the members of the group have already been unjustly disadvantaged. I assess this reasoning. First, I distinguish the anti-compounding duty from a related but distinct duty—the proportionality duty—from which at least some of the intuitive appeal of the (...) former illegitimately derives. Second, I distinguish between different versions of the anti-compounding duty, arguing that, on some versions, uses of algorithm-assisted decision procedures rarely clash with the anti-compounding duty. Third, drawing on examples of algorithm-assisted decision procedures, I present three objections to the idea that there is a reason not to compound injustice. The most important of these is that one can compound injustice in a non-disrespectful way, and that the wrongfulness of non-disrespectfully compounding injustice is fully explained by the proportionality duty. (shrink)

With the increasing use of algorithms in high-stakes areas such as criminal justice and health has come a significant concern about the fairness of prediction-based decision procedures. In this article I argue that a prominent class of mathematically incompatible performance parity criteria can all be understood as applications of John Broome’s account of fairness as the proportional satisfaction of claims. On this interpretation these criteria do not disagree on what it means for an algorithm to be _fair_. Rather they express (...) different understandings of what grounds a claim to a good being allocated by an algorithmic decision procedure. I then argue that an important implication of the Broomean interpretation is that it strengthens the case for outcome-based criteria. Finally, I consider how a version of the levelling-down objection to performance parity criteria arises within the Broomean account. (shrink)

Künstliche Intelligenz (KI) ist heute schon ein fester Bestandteil unseres Lebens, auch wenn sie oft im Verborgenen wirkt. Wo wird diese Entwicklung hinführen und was wird das für uns bedeuten? Jens Kipper erklärt, wie moderne KI funktioniert, was sie heute schon kann und welche Auswirkungen ihre Verwendung in Waffensystemen, in der Medizin und Wissenschaft, im Arbeitsleben und anderswo haben wird. Kipper argumentiert dafür, dass die Entwicklung von KI zu großen gesellschaftlichen Umwälzungen führen wird. Er erläutert zudem, wovon es abhängt, dass (...) diese Umwälzungen erfreulicher Natur sein werden und was wir tun können, um die durch den Einsatz von KI geprägten Veränderungen in die richtige Richtung zu lenken. (shrink)

The increasing implementation of and reliance on machine-learning (ML) algorithms to perform tasks, deliver services and make decisions in health and healthcare have made the need for fairness in ML, and more specifically in healthcare ML algorithms (HMLA), a very important and urgent task. However, while the debate on fairness in the ethics of artificial intelligence (AI) and in HMLA has grown significantly over the last decade, the very concept of fairness as an ethical value has not yet been sufficiently (...) explored. Our paper aims to fill this gap and address the AI ethics principle of fairness from a conceptual standpoint, drawing insights from accounts of fairness elaborated in moral philosophy and using them to conceptualise fairness as an ethical value and to redefine fairness in HMLA accordingly. To achieve our goal, following a first section aimed at clarifying the background, methodology and structure of the paper, in the second section, we provide an overview of the discussion of the AI ethics principle of fairness in HMLA and show that the concept of fairness underlying this debate is framed in purely distributive terms and overlaps with non-discrimination, which is defined in turn as the absence of biases. After showing that this framing is inadequate, in the third section, we pursue an ethical inquiry into the concept of fairness and argue that fairness ought to be conceived of as an ethical value. Following a clarification of the relationship between fairness and non-discrimination, we show that the two do not overlap and that fairness requires much more than just non-discrimination. Moreover, we highlight that fairness not only has a distributive but also a socio-relational dimension. Finally, we pinpoint the constitutive components of fairness. In doing so, we base our arguments on a renewed reflection on the concept of respect, which goes beyond the idea of equal respect to include respect for individual persons. In the fourth section, we analyse the implications of our conceptual redefinition of fairness as an ethical value in the discussion of fairness in HMLA. Here, we claim that fairness requires more than non-discrimination and the absence of biases as well as more than just distribution; it needs to ensure that HMLA respects persons both as persons and as particular individuals. Finally, in the fifth section, we sketch some broader implications and show how our inquiry can contribute to making HMLA and, more generally, AI promote the social good and a fairer society. (shrink)

Machine learning classifiers are increasingly used to inform, or even make, decisions significantly affecting human lives. Fairness concerns have spawned a number of contributions aimed at both identifying and addressing unfairness in algorithmic decision-making. This paper critically discusses the adoption of group-parity criteria (e.g., demographic parity, equality of opportunity, treatment equality) as fairness standards. To this end, we evaluate the use of machine learning methods relative to different steps of the decision-making process: assigning a predictive score, linking a classification to (...) the score, and adopting decisions based on the classification. Throughout our inquiry we use the COMPAS system, complemented by a radical simplification of it (our SAPMOC I and SAPMOC II models), as our running examples. Through these examples, we show how a system that is equally accurate for different groups may fail to comply with group-parity standards, owing to different base rates in the population. We discuss the general properties of the statistics determining the satisfaction of group-parity criteria and levels of accuracy. Using the distinction between scoring, classifying, and deciding, we argue that equalisation of classifications/decisions between groups can be achieved thorough group-dependent thresholding. We discuss contexts in which this approach may be meaningful and useful in pursuing policy objectives. We claim that the implementation of group-parity standards should be left to competent human decision-makers, under appropriate scrutiny, since it involves discretionary value-based political choices. Accordingly, predictive systems should be designed in such a way that relevant policy goals can be transparently implemented. Our paper presents three main contributions: (1) it addresses a complex predictive system through the lens of simplified toy models; (2) it argues for selective policy interventions on the different steps of automated decision-making; (3) it points to the limited significance of statistical notions of fairness to achieve social goals. (shrink)

Recent work in AI ethics often calls for AI to advance human values and interests. The concept of “AI for people” is one notable example. Though commendable in some respects, this work falls short by excluding the moral significance of nonhumans. This paper calls for a shift in AI ethics to more inclusive paradigms such as “AI for the world” and “AI for the universe”. The paper outlines the case for more inclusive paradigms and presents implications for moral philosophy and (...) computer science work on AI ethics. (shrink)

The use of AI algorithms in criminal trials has been the subject of very lively ethical and legal debates recently. While there are concerns over the lack of accuracy and the harmful biases that certain algorithms display, new algorithms seem more promising and might lead to more accurate legal decisions. Algorithms seem especially relevant for bail decisions, because such decisions involve statistical data to which human reasoners struggle to give adequate weight. While getting the right legal outcome is a strong (...) desideratum of criminal trials, advocates of the relational theory of procedural justice give us good reason to think that fairness and perceived fairness of legal procedures have a value that is independent from the outcome. According to this literature, one key aspect of fairness is trustworthiness. In this paper, I argue that using certain algorithms to assist bail decisions could increase three diferent aspects of judges’ trustworthiness: (1) actual trustworthiness, (2) rich trustworthiness, and (3) perceived trustworthiness. (shrink)

Philosophy &Public Affairs, Volume 51, Issue 2, Page 166-190, Spring 2023.

Can the evidence provided by software systems meet the standard of proof for civil or criminal cases, and is it individualized evidence? Or, to the contrary, do software systems exclusively provide bare statistical evidence? In this paper, we argue that there are cases in which evidence in the form of probabilities computed by software systems is not bare statistical evidence, and is thus able to meet the standard of proof. First, based on the case of State v. Loomis, we investigate (...) recidivism predictions provided by software systems used in the courtroom. Here, we raise problems for software systems that provide predictions that are based on bare statistical evidence. Second, by examining the case of People v. Chubbs, we argue that the statistical evidence provided by software systems in cold hit DNA cases may in some cases suffice for individualized evidence, on a view on which individualized evidence is evidence that normically supports the relevant proposition (Smith, in Mind 127:1193–1218, 2018). (shrink)

Generative artificial intelligence systems based on transformers, including both text-generators like GPT-3 and image generators like DALL-E 2, have recently entered the popular consciousness. These tools, while impressive, are liable to reproduce, exacerbate, and reinforce extant human social biases, such as gender and racial biases. In this paper, we systematically review the extent to which DALL-E Mini suffers from this problem. In line with the Model Card published alongside DALL-E Mini by its creators, we find that the images it produces (...) tend to represent dozens of different occupations as populated either solely by men (e.g., pilot, builder, plumber) or solely by women (e.g., hairdresser, receptionist, dietitian). In addition, the images DALL-E Mini produces tend to represent most occupations as populated primarily or solely by White people (e.g., farmer, painter, prison officer, software engineer) and very few by non-White people (e.g., pastor, rapper). These findings suggest that exciting new AI technologies should be critically scrutinized and perhaps regulated before they are unleashed on society. (shrink)

Statistical criteria of fairness are formal measures of how an algorithm performs that aim to help us determine whether an algorithm would be fair to use in decision-making. In this paper, I introduce a new version of the criterion known as “Equalized Odds,” argue that it is a requirement of procedural fairness, and show that it is immune to a number of objections to the standard version.

Racial, Gender, and Intersectional Biases in AI / -/- Dominant View of Intersectional Fairness in the AI Literature / -/- Three Fundamental Problems with the Dominant View / 1. Overemphasis on Intersections of Attributes / 2. Dilemma between Infinite Regress and Fairness Gerrymandering / 3. Narrow Understanding of Fairness as Parity / -/- Rethinking AI Fairness: from Weak to Strong Fairness.

A growing number of studies on fairness in artificial intelligence (AI) use the notion of intersectionality to measure AI fairness. Most of these studies take intersectional fairness to be a matter of statistical parity among intersectional subgroups: an AI algorithm is “intersectionally fair” if the probability of the outcome is roughly the same across all subgroups defined by different combinations of the protected attributes. This paper identifies and examines three fundamental problems with this dominant interpretation of intersectional fairness in AI. (...) First, the dominant approach is so preoccupied with the intersection of attributes/categories (e.g., race, gender) that it fails to address the intersection of oppression (e.g., racism, sexism), which is more central to intersectionality as a critical framework. Second, the dominant approach faces a dilemma between infinite regress and fairness gerrymandering: it either keeps splitting groups into smaller subgroups or arbitrarily selects protected groups. Lastly, the dominant view fails to capture what it really means for AI algorithms to be fair, in terms of both distributive and non-distributive fairness. I distinguish a strong sense of AI fairness from a weak sense that is prevalent in the literature, and conclude by envisioning paths towards strong intersectional fairness in AI. (shrink)

Should machines be judges? Some say no, arguing that citizens would see robot-led legal proceedings as procedurally unfair because “having your day in court” is having another human adjudicate your claims. Prior research established that people obey the law in part because they see it as procedurally just. The introduction of artificially intelligent (AI) judges could therefore undermine sentiments of justice and legal compliance if citizens intuitively take machine-adjudicated proceedings to be less fair than the human-adjudicated status quo. Two original (...) experiments show that ordinary people share this intuition. There is a perceived “human-AI fairness gap.” -/- However, it is also possible to reduce — and perhaps even eliminate — the fairness gap through “algorithmic offsetting.” Affording a hearing before AI judges and enhancing the interpretability of AI-rendered decisions reduce the human-AI fairness gap. Moreover, the procedural justice advantage of a human over AI appears to be driven more by beliefs about the accuracy of the outcome and thoroughness of consideration, rather than doubts about whether a party felt it had a good opportunity to voice its opinions or whether the judge understood the perspective of the litigant. -/- The results support a common and fundamental objection to robot judges: There is a concerning human-AI fairness gap. Yet, the results also indicate that the strongest version of this challenge — human judges have irreducible procedural fairness advantages — is not reflected in public views. In some circumstances, people see a day in robot court as no less fair than a day in human court. (shrink)

This note contains a corrective and a generalization of results by Borsboom et al. (2008), based on Heesen and Romeijn (2019). It highlights the relevance of insights from psychometrics beyond the context of psychological testing.

In this paper I argue that some examples of what we label ‘algorithmic bias’ would be better understood as cases of institutional bias. Even when individual algorithms appear unobjectionable, they may produce biased outcomes given the way that they are embedded in the background structure of our social world. Therefore, the problematic outcomes associated with the use of algorithmic systems cannot be understood or accounted for without a kind of structural account. Understanding algorithmic bias as institutional bias in particular (as (...) opposed to other structural accounts) has at least two important upshots. First, I argue that the existence of bias that is intrinsic to certain institutions (whether algorithmic or not) suggests that at least in some cases, the algorithms now substituting as pieces of institutional norms or rules are not “fixable” in the relevant sense, because the institutions they help make up are not fixable. Second, I argue that in other cases, changing the algorithms being used within our institutions (rather than getting rid of them entirely) is essential to changing the background structural conditions of our society. (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)

What does it mean for algorithmic classifications to be fair to different socially salient groups? According to classification parity criteria, what is required is equality across groups with respect to some performance measure such as error rates. Critics of classification parity object that classification parity entails that achieving fairness may require us to choose an algorithm that makes no group better off and some groups worse off than an alternative. In this article, I interpret the problem of algorithmic fairness as (...) a case concerning the ethics of the distribution of algorithmic classifications across groups (as opposed to, e.g., the fairness of data collection). I begin with a short introduction of algorithmic fairness as a problem discussed in machine learning. I then show how the criticism raised against classification parity is a form of leveling down objection, and I interpret the egalitarianism of classification parity as deontic egalitarianism. I then discuss a challenge to this interpretation and suggest a revision. Finally, I examine how my interpretation provides proponents of classification parity with a response to the leveling down criticism and how it relates to a recent suggestion to evaluate fairness for automated decision-making systems based on risk and welfare considerations from behind a veil of ignorance. (shrink)

Algorithmic discrimination has rapidly become a topic of intense public and academic interest. This article explores three issues raised by algorithmic discrimination: 1) the distinction between direct and indirect discrimination, 2) the notion of disadvantageous treatment, and 3) the moral badness of discriminatory automated decision-making. It argues that some conventional distinctions between direct and indirect discrimination appear not to apply to algorithmic discrimination, that algorithmic discrimination may often be discrimination between groups, as opposed to against groups, and that it is (...) not necessarily the case that morally bad algorithmic discrimination gives us reason to not use automated decision-making. For each of the three issues, the article explores implications for algorithmic discrimination, suggests some alternative answers, and clarifies how we may want to think of discrimination more broadly in light of lessons drawn from the context of algorithmic discrimination. (shrink)

Machine learning (ML) techniques have become pervasive across a range of different applications, and are now widely used in areas as disparate as recidivism prediction, consumer credit-risk analysis, and insurance pricing. Likewise, in the physical world, ML models are critical components in autonomous agents such as robotic surgeons and self-driving cars. Among the many ethical dimensions that arise in the use of ML technology in such applications, analyzing morally permissible actions is both immediate and profound. For example, there is the (...) potential for learned algorithms to become biased against certain groups. More generally, in so much that the decisions of ML models impact society, both virtually (e.g., denying a loan) and physically (e.g., driving into a pedestrian), notions of accountability, blame and responsibility need to be carefully considered. In this article, we advocate for a two-pronged approach ethical decision-making enabled using rich models of autonomous agency: on the one hand, we need to draw on philosophical notions of such as beliefs, causes, effects and intentions, and look to formalise them, as attempted by the knowledge representation community, but on the other, from a computational perspective, such theories need to also address the problems of tractable reasoning and (probabilistic) knowledge acquisition. As a concrete instance of this tradeoff, we report on a few preliminary results that apply (propositional) tractable probabilistic models to problems in fair ML and automated reasoning of moral principles. Such models are compilation targets for certain types of knowledge representation languages, and can effectively reason in service some computational tasks. They can also be learned from data. Concretely, current evidence suggests that they are attractive structures for jointly addressing three fundamental challenges: reasoning about possible worlds + tractable computation + knowledge acquisition. Thus, these seems like a good starting point for modelling reasoning robots as part of the larger ecosystem where accountability and responsibility is understood more broadly. (shrink)

Bias infects the algorithms that wield increasing control over our lives. Predictive policing systems overestimate crime in communities of color; hiring algorithms dock qualified female candidates; and facial recognition software struggles to recognize dark-skinned faces. Algorithmic bias has received significant attention. Algorithmic neutrality, in contrast, has been largely neglected. Algorithmic neutrality is my topic. I take up three questions. What is algorithmic neutrality? Is algorithmic neutrality possible? When we have algorithmic neutrality in mind, what can we learn about algorithmic bias? (...) To answer these questions in concrete terms, I work with a case study: search engines. Drawing on work about neutrality in science, I say that a search engine is neutral only if certain values—like political ideologies or the financial interests of the search engine operator—play no role in how the search engine ranks pages. Search neutrality, I argue, is impossible. Its impossibility seems to threaten the significance of search bias: if no search engine is neutral, then every search engine is biased. To defuse this threat, I distinguish two forms of bias—failing-on-its-own-terms bias and exogenous-values bias. This distinction allows us to make sense of search bias, and capture its normative complexion, despite the impossibility of neutrality. (shrink)

Data-driven predictive algorithms are widely used to automate and guide high-stake decision making such as bail and parole recommendation, medical resource distribution, and mortgage allocation. Nevertheless, harmful outcomes biased against vulnerable groups have been reported. The growing research field known as 'algorithmic fairness' aims to mitigate these harmful biases. Its primary methodology consists in proposing mathematical metrics to address the social harms resulting from an algorithm's biased outputs. The metrics are typically motivated by -- or substantively rooted in -- ideals (...) of distributive justice, as formulated by political and legal philosophers. The perspectives of feminist political philosophers on social justice, by contrast, have been largely neglected. Some feminist philosophers have criticized the local scope of the paradigm of distributive justice and have proposed corrective amendments to surmount its limitations. The present paper brings some key insights of feminist political philosophy to algorithmic fairness. The paper has three goals. First, I show that algorithmic fairness does not accommodate structural injustices in its current scope. Second, I defend the relevance of structural injustices -- as pioneered in the contemporary philosophical literature by Iris Marion Young -- to algorithmic fairness. Third, I take some steps in developing the paradigm of 'responsible algorithmic fairness' to correct for errors in the current scope and implementation of algorithmic fairness. I close by some reflections of directions for future research. (shrink)

This paper presents an intuitive explanation about why and how Rawlsian Theory of Justice (Rawls in A theory of justice, Harvard University Press, Harvard, 1971) provides the foundations to a solution for algorithmic bias. The contribution of the paper is to discuss and show why Rawlsian ideas in their original form (e.g. the veil of ignorance, original position, and allowing inequalities that serve the worst-off) are relevant to operationalize fairness for algorithmic decision making. The paper also explains how this leads (...) to a specific MinMaxfairness solution, which addresses the basic challenges of algorithmic justice. We combine substantive elements of Rawlsian perspective with an intuitive explanation in order to provide accessible and practical insights. The goal is to propose and motivate why and how the MinMaxfairness solution derived from Rawlsian principles overcomes some of the current challenges for algorithmic bias and highlight the benefits provided when compared to other approaches. The paper presents and discusses the solution by building a bridge between the qualitative theoretical aspects and the quantitative technical approach. (shrink)

Artificially intelligent (AI) technologies are increasingly being used in many workplaces. It is recognised that there are ethical dimensions to the ways in which organisations implement AI alongside, or substituting for, their human workforces. How will these technologically driven disruptions impact the employee–employer exchange? We provide one way to explore this question by drawing on scholarship linking Integrative Social Contracts Theory (ISCT) to the psychological contract (PC). Using ISCT, we show that the macrosocial contract’s ethical AI norms of beneficence, non-maleficence, (...) autonomy, justice, and explicability can feed into national- and organisational-level microsocial contracts. We also outline the role of employees’ technology frames in this process. We then use an illustrative example to demonstrate how this multilevel normative background helps to inform the content of individuals’ PCs in the context of working with AI technologies. (shrink)

BackgroundThere is a growing concern about artificial intelligence (AI) applications in healthcare that can disadvantage already under-represented and marginalised groups (eg, based on gender or race).ObjectivesOur objectives are to canvas the range of strategies stakeholders endorse in attempting to mitigate algorithmic bias, and to consider the ethical question of responsibility for algorithmic bias.MethodologyThe study involves in-depth, semistructured interviews with healthcare workers, screening programme managers, consumer health representatives, regulators, data scientists and developers.ResultsFindings reveal considerable divergent views on three key issues. First, (...) views on whether bias is a problem in healthcare AI varied, with most participants agreeing bias is a problem (which we call the bias-critical view), a small number believing the opposite (the bias-denial view), and some arguing that the benefits of AI outweigh any harms or wrongs arising from the bias problem (the bias-apologist view). Second, there was a disagreement on the strategies to mitigate bias, and who is responsible for such strategies. Finally, there were divergent views on whether to include or exclude sociocultural identifiers (eg, race, ethnicity or gender-diverse identities) in the development of AI as a way to mitigate bias.Conclusion/significanceBased on the views of participants, we set out responses that stakeholders might pursue, including greater interdisciplinary collaboration, tailored stakeholder engagement activities, empirical studies to understand algorithmic bias and strategies to modify dominant approaches in AI development such as the use of participatory methods, and increased diversity and inclusion in research teams and research participant recruitment and selection. (shrink)

Artificial intelligence (AI) systems are increasingly adopted to make decisions in domains such as business, education, health care, and criminal justice. However, such algorithmic decision systems can have prevalent biases against marginalized social groups and undermine social justice. Explainable artificial intelligence (XAI) is a recent development aiming to make an AI system’s decision processes less opaque and to expose its problematic biases. This paper argues against technical XAI, according to which the detection and interpretation of algorithmic bias can be handled (...) more or less independently by technical experts who specialize in XAI methods. Drawing on resources from feminist epistemology, we show why technical XAI is mistaken. Specifically, we demonstrate that the proper detection of algorithmic bias requires relevant interpretive resources, which can only be made available, in practice, by actively involving a diverse group of stakeholders. Finally, we suggest how feminist theories can help shape integrated XAI: an inclusive social-epistemic process that facilitates the amelioration of algorithmic bias. (shrink)

Feminized AIs designed for in-home verbal assistance are often subjected to gendered verbal abuse by their users. I survey a variety of features contributing to this phenomenon—from financial incentives for businesses to build products likely to provoke gendered abuse, to the impact of such behavior on household members—and identify a potential worry for attempts to criticize the phenomenon; while critics may be tempted to argue that engaging in gendered abuse of AI increases the chances that one will direct this abuse (...) toward human beings, the recent history of attempts to connect video game violence to real-world aggression suggests that things may not be so simple. I turn to Confucian discussions of the role of ritualized social interactions both to better understand the roots of the problem and to investigate potential strategies for improvement, given a complex interplay between designers and device users. I argue that designers must grapple with the entrenched sexism in our society, at the expense of “smooth” and “seamless” user interfaces, in order to intentionally disrupt entrenched but harmful patterns of interaction, but that doing so is both consistent with and recommended by Confucian accounts of social rituals. (shrink)

Over the past decade, feminist philosophers have gone a long way toward identifying and explaining the phenomenon that has come to be known as epistemic injustice. Epistemic injustice is injustice occurring within the domain of knowledge (e.g., knowledge production and transmission), which typically impacts structurally marginalized social groups. In this paper, we argue that, as they currently work, algorithms on social media exacerbate the problem of epistemic injustice and related problems of social distrust. In other words, we argue that algorithms (...) on social media recreate and reify the conditions that lead to some groups being systematically denied the full status of knowers, thereby corrupting the epistemic terrain and, with it, systems of social trust and cooperation. We argue that algorithms do this in two ways—namely, via what we are calling algorithmic targeting and algorithmic sorting. (shrink)

We argue that machine learning algorithms can inflict microaggressions on members of marginalized groups and that recognizing these harms as instances of microaggressions is key to effectively addressing the problem. The concept of microaggression is also illuminated by being studied in algorithmic contexts. We contribute to the microaggression literature by expanding the category of environmental microaggressions and highlighting the unique issues of moral responsibility that arise when we focus on this category. We theorize two kinds of algorithmic microaggression, stereotyping and (...) erasure microaggressions, and argue that corporations are responsible for the microaggressions their algorithms create. As a case study, we look at the problems faced by Google’s autocomplete prediction and at the insufficiency of their solutions. The case study of autocomplete demonstrates our core claim that microaggressions constitute a distinct form of algorithmic bias and that identifying them as such allows us to avoid seeming solutions that recreate the same kinds of harms. Google has a responsibility to make information freely available, without exposing users to degradation. To fulfill its duties to marginalized groups, Google must abandon the fiction of neutral prediction and instead embrace the liberatory power of suggestion. (shrink)

Algorithmic fairness is a new interdisciplinary field of study focused on how to measure whether a process, or algorithm, may unintentionally produce unfair outcomes, as well as whether or how the potential unfairness of such processes can be mitigated. Statistical discrimination describes a set of informational issues that can induce rational (i.e., Bayesian) decision-making to lead to unfair outcomes even in the absence of discriminatory intent. In this article, we provide overviews of these two related literatures and draw connections between (...) them. The comparison illustrates both the conflict between rationality and fairness and the importance of endogeneity (e.g., “rational expectations” and “self-fulfilling prophecies”) in defining and pursuing fairness. Taken in concert, we argue that the two traditions suggest a value for considering new fairness notions that explicitly account for how the individual characteristics an algorithm intends to measure may change in response to the algorithm. (shrink)

It is increasingly evident that if researchers and policymakers want to meaningfully develop an understanding of responsible innovation, we must first ask whether some sociotechnical systems should be developed, at all. Here I argue that systems like facial recognition, predictive policing, and biometrics are predicated on myriad human prejudicial biases and assumptions which must be named and interrogated prior to any innovation. Further, the notions of individual responsibility inherent in discussions of technological ethics and fairness overburden marginalized peoples with a (...) demand to prove the reality of their marginalization. Instead, we should focus on equity and justice, valuing the experiential knowledge of marginalized peoples and optimally positioning them to enact deep, lasting change. My position aligns with those in Science, Technology, and Society (STS) which center diverse and situated knowledges, and is articulated together with calls for considering within science and engineering wider sociocultural concerns like justice and equality. (shrink)

Uncovering the world’s ethnic inequalities is hampered by a lack of ethnicity-annotated datasets. Name-ethnicity classifiers (NECs) can help, as they are able to infer people’s ethnicities from their names. However, since the latest generation of NECs rely on machine learning and artificial intelligence (AI), they may suffer from the same racist and sexist biases found in many AIs. Therefore, this paper offers an algorithmic fairness audit of three NECs. It finds that the UK-Census-trained EthnicityEstimator displays large accuracy biases with regards (...) to ethnicity, but relatively less among gender and age groups. In contrast, the Twitter-trained NamePrism and the Wikipedia-trained Ethnicolr are more balanced among ethnicity, but less among gender and age. We relate these biases to global power structures manifested in naming conventions and NECs’ input distribution of names. To improve on the uncovered biases, we program a novel NEC, N2E, using fairness-aware AI techniques. We make N2E freely available at www.name-to-ethnicity.com. (shrink)

Google Search is arguably one of the primary epistemic tools in use today, with the lion’s share of the search-engine market globally. Scholarship on countering the current scourge of misinformation often recommends “digital lit- eracy” where internet users, especially those who get their information from so- cial media, are encouraged to fact-check such information using reputable sources. Given our current internet-based epistemic landscape, and Google’s dominance of the internet, it is very likely that such acts of epistemic hygiene will take (...) place via Google Search. The question arises whether Google Search is fit for purpose, given the apparent misalignment the general epistemic goal of promoting true beliefs and the greater online commercial ecosystem in which it is embedded. I argue that Google Search is epistemically problematic as it stands, mainly due to the opacity related to the parameters it uses for personalising search results. I further argue that in as far as an ordinary internet user is legitimately ignorant of Google’s workings, uses it in an “ordinary manner”, and is generally unable to avoid using it in the current information environment, they are not ep- istemically blameworthy for any false beliefs that they acquire via it. I conclude that too much emphasis is currently placed on individual epistemic practices and not enough on our information environment and epistemic tools when it comes to countering misinformation. (shrink)

A self-fulfilling prophecy is, roughly, a prediction that brings about its own truth. Although true predictions are hard to fault, self-fulfilling prophecies are often regarded with suspicion. In this article, we vindicate this suspicion by explaining what self-fulfilling prophecies are and what is problematic about them, paying special attention to how their problems are exacerbated through automated prediction. Our descriptive account of self-fulfilling prophecies articulates the four elements that define them. Based on this account, we begin our critique by showing (...) that typical self-fulfilling prophecies arise due to mistakes about the relationship between a prediction and its object. Such mistakes—along with other mistakes in predicting or in the larger practical endeavor—are easily overlooked when the predictions turn out true. Thus we note that self-fulfilling prophecies prompt no error signals; truth shrouds their mistakes from humans and machines alike. Consequently, self-fulfilling prophecies create several obstacles to accountability for the outcomes they produce. We conclude our critique by showing how failures of accountability, and the associated failures to make corrections, explain the connection between self-fulfilling prophecies and feedback loops. By analyzing the complex relationships between accuracy and other evaluatively significant features of predictions, this article sheds light both on the special case of self-fulfilling prophecies and on the ethics of prediction more generally. (shrink)

Over the past decade, scholars, institutions, and activists have voiced strong concerns about the potential of automated decision systems to indirectly discriminate against vulnerable groups. This article analyses the ethics of algorithmic indirect discrimination, and argues that we can explain what is morally bad about such discrimination by reference to the fact that it causes harm. The article first sketches certain elements of the technical and conceptual background, including definitions of direct and indirect algorithmic differential treatment. It next introduces three (...) prominent accounts of fairness as potential explanations if the badness of algorithmic indirect discrimination, but argues that all three are vulnerable to powerful levelling-down-style objections. Instead, the article demonstrates how proper attention to the way differences in decision-scenarios affect the distribution of harms can help us account for intuitions in prominent cases. Finally, the article considers a potential objection based on the fact that certain forms of algorithmic indirect discrimination appear to distribute rather than cause harm, and notes that we can explain how such distributions cause harm by attending to differences in individual and group vulnerability. (shrink)