In the age of Big Data, companies and governments are increasingly using algorithms to inform hiring decisions, employee management, policing, credit scoring, insurance pricing, and many more aspects of our lives. Artificial intelligence systems can help us make evidence-driven, efficient decisions, but can also confront us with unjustified, discriminatory decisions wrongly assumed to be accurate because they are made automatically and quantitatively. It is becoming evident that these technological developments are consequential to people’s fundamental human rights. Despite increasing attention to (...) these urgent challenges in recent years, technical solutions to these complex socio-ethical problems are often developed without empirical study of societal context and the critical input of societal stakeholders who are impacted by the technology. On the other hand, calls for more ethically and socially aware AI often fail to provide answers for how to proceed beyond stressing the importance of transparency, explainability, and fairness. Bridging these socio-technical gaps and the deep divide between abstract value language and design requirements is essential to facilitate nuanced, context-dependent design choices that will support moral and social values. In this paper, we bridge this divide through the framework of Design for Values, drawing on methodologies of Value Sensitive Design and Participatory Design to present a roadmap for proactively engaging societal stakeholders to translate fundamental human rights into context-dependent design requirements through a structured, inclusive, and transparent process. (shrink)

This article provides a conceptual map of the debate on ideal and non‐ideal theory. It argues that this debate encompasses a number of different questions, which have not been kept sufficiently separate in the literature. In particular, the article distinguishes between the following three interpretations of the ‘ideal vs. non‐ideal theory’ contrast: (i) full compliance vs. partial compliance theory; (ii) utopian vs. realistic theory; (iii) end‐state vs. transitional theory. The article advances critical reflections on each of these sub‐debates, and highlights (...) areas for future research in the field. (shrink)

Recently, there has been a large amount of support for the idea that causal claims can be sensitive to normative considerations. Previous work has focused on the concept of actual causation, defending the claim that whether or not some token event c is a cause of another token event e is influenced by both statistical and prescriptive norms. I focus on the policy debate surrounding alternative energies, and use the causal modelling framework to show that in this context, people’s normative (...) commitments don’t just affect the causal claims they are willing to endorse, but also their understanding of the causal structure. In the context of the alternative energy debate, normative considerations affect our understanding of the causal structure by influencing our judgements about which variables should be held fixed, and therefore which variables should be relegated to the background of a causal model. In cases of extreme disagreement, normative commitments can also affect which causal structure we think should be instantiated. Thus, focusing on a new context has revealed a previously unexplored sense in which normative factors are incorporated into causal reasoning. (shrink)

There is widespread agreement among both supporters and opponents that affirmative action either must not violate any principle of equal opportunity or procedural justice, or if it does, it may do so only given current extenuating circumstances. Many believe that affirmative action is morally problematic, only justified to the extent that it brings us closer to the time when we will no longer need it. In other words, those that support affirmative action believe it is acceptable in nonideal theory, but (...) not ideal theory. This paper argues that affirmative action is entirely compatible with equal opportunity and procedural justice and would be even in an ideal world. I defend a new analysis of Rawlsian procedural justice according to which it is permissible to interfere in the outcomes of procedures, and thus I show that affirmative action is not morally problematic in the way that many have supposed. (shrink)

Existing research suggests that people's judgments of actual causation can be influenced by the degree to which they regard certain events as normal. We develop an explanation for this phenomenon that draws on standard tools from the literature on graphical causal models and, in particular, on the idea of probabilistic sampling. Using these tools, we propose a new measure of actual causal strength. This measure accurately captures three effects of normality on causal judgment that have been observed in existing studies. (...) More importantly, the measure predicts a new effect ("abnormal deflation"). Two studies show that people's judgments do, in fact, show this new effect. Taken together, the patterns of people's causal judgments thereby provide support for the proposed explanation. (shrink)

Data‐driven algorithms are widely used to make or assist decisions in sensitive domains, including healthcare, social services, education, hiring, and criminal justice. In various cases, such algorithms have preserved or even exacerbated biases against vulnerable communities, sparking a vibrant field of research focused on so‐called algorithmic biases. This research includes work on identification, diagnosis, and response to biases in algorithm‐based decision‐making. This paper aims to facilitate the application of philosophical analysis to these contested issues by providing an overview of three (...) key topics: What is algorithmic bias? Why and how can it occur? What can and should be done about it? Throughout, we highlight connections—both actual and potential—with philosophical ideas and concerns. (shrink)

Th is paper investigates the epistemic powers of democratic institutions through an assessment of three epistemic models of democracy : the Condorcet Jury Th eorem, the Diversity Trumps Ability Th eorem, and Dewey's experimentalist model. Dewey's model is superior to the others in its ability to model the epistemic functions of three constitutive features of democracy : the epistemic diversity of participants, the interaction of voting with discussion, and feedback mechanisms such as periodic elections and protests. It views democracy as (...) an institution for pooling widely distributed information about problems and policies of public interest by engaging the participation of epistemically diverse knowers. Democratic norms of free discourse, dissent, feedback, and accountability function to ensure collective, experimentallybased learning from the diverse experiences of diff erent knowers. I illustrate these points with a case study of community forestry groups in South Asia, whose epistemic powers have been hobbled by their suppression of women's participation. (shrink)

We map the recently proposed notions of algorithmic fairness to economic models of Equality of opportunity (EOP)---an extensively studied ideal of fairness in political philosophy. We formally show that through our conceptual mapping, many existing definition of algorithmic fairness, such as predictive value parity and equality of odds, can be interpreted as special cases of EOP. In this respect, our work serves as a unifying moral framework for understanding existing notions of algorithmic fairness. Most importantly, this framework allows us to (...) explicitly spell out the moral assumptions underlying each notion of fairness, and interpret recent fairness impossibility results in a new light. Last but not least and inspired by luck egalitarian models of EOP, we propose a new family of measures for algorithmic fairness. We illustrate our proposal empirically and show that employing a measure of algorithmic (un)fairness when its underlying moral assumptions are not satisfied, can have devastating consequences for the disadvantaged group's welfare. (shrink)