Responsible AI (RAI) guidelines aim to ensure that AI systems respect democratic values. While a step in the right direction, they currently fail to impact practice. Our work discusses reasons for this lack of impact and clusters them into five areas: (1) the abstract nature of RAI guidelines, (2) the problem of selecting and reconciling values, (3) the difficulty of operationalising RAI success metrics, (4) the fragmentation of the AI pipeline, and (5) the lack of internal advocacy and accountability. Afterwards, (...) we introduce a number of approaches to RAI from a range of disciplines, exploring their potential as solutions to the identified challenges. We anchor these solutions in practice through concrete examples, bridging the gap between the theoretical considerations of RAI and on-the-ground processes that currently shape how AI systems are built. Our work considers the socio-technical nature of RAI limitations and the resulting necessity of producing socio-technical solutions. (shrink)

The need for fair and just AI is often related to the possibility of understanding AI itself, in other words, of turning an opaque box into a glass box, as inspectable as possible. Transparency and explainability, however, pertain to the technical domain and to philosophy of science, thus leaving the ethics and epistemology of AI largely disconnected. To remedy this, we propose an integrated approach premised on the idea that a glass-box epistemology should explicitly consider how to incorporate values and (...) other normative considerations, such as intersectoral vulnerabilities, at critical stages of the whole process from design and implementation to use and assessment. To connect ethics and epistemology of AI, we perform a double shift of focus. First, we move from trusting the output of an AI system to trusting the process that leads to the outcome. Second, we move from expert assessment to more inclusive assessment strategies, aiming to facilitate expert and non-expert assessment. Together, these two moves yield a framework usable for experts and non-experts when they inquire into relevant epistemological and ethical aspects of AI systems. We dub our framework ‘epistemology-cum-ethics’ to signal the equal importance of both aspects. We develop it from the vantage point of the designers: how to create the conditions to internalize values into the whole process of design, implementation, use, and assessment of an AI system, in which values (epistemic and non-epistemic) are explicitly considered at each stage and inspectable by every salient actor involved at any moment. (shrink)

Algorithmic systems are increasingly used by state agencies to inform decisions about humans. They produce scores on risks of recidivism in criminal justice, indicate the probability for a job seeker to find a job in the labor market, or calculate whether an applicant should get access to a certain university program. In this contribution, we take an interdisciplinary perspective, provide a bird’s eye view of the different key decisions that are to be taken when state actors decide to use an (...) algorithmic system, and illustrate these decisions with empirical examples from case studies. Building on these insights, we discuss the main pitfalls and promises of the use of algorithmic system by the state and focus on four levels: The most basic question whether an algorithmic system should be used at all, the regulation and governance of the system, issues of algorithm design, and, finally, questions related to the implementation of the system on the ground and the human–machine-interaction that comes with it. Based on our assessment of the advantages and challenges that arise at each of these levels, we propose a set of crucial questions to be asked when such intricate matters are addressed. (shrink)

Efforts to promote equitable public policy with algorithms appear to be fundamentally constrained by the “impossibility of fairness” (an incompatibility between mathematical definitions of fairness). This technical limitation raises a central question about algorithmic fairness: How can computer scientists and policymakers support equitable policy reforms with algorithms? In this article, I argue that promoting justice with algorithms requires reforming the methodology of algorithmic fairness. First, I diagnose the problems of the current methodology for algorithmic fairness, which I call “formal algorithmic (...) fairness.” Because formal algorithmic fairness restricts analysis to isolated decision-making procedures, it leads to the impossibility of fairness and to models that exacerbate oppression despite appearing “fair.” Second, I draw on theories of substantive equality from law and philosophy to propose an alternative methodology, which I call “substantive algorithmic fairness.” Because substantive algorithmic fairness takes a more expansive scope of analysis, it enables an escape from the impossibility of fairness and provides a rigorous guide for alleviating injustice with algorithms. In sum, substantive algorithmic fairness presents a new direction for algorithmic fairness: away from formal mathematical models of “fair” decision-making and toward substantive evaluations of whether and how algorithms can promote justice in practice. (shrink)

Modern society makes extensive use of automated algorithmic decisions, fueled by advances in artificial intelligence. However, since these systems are not perfect, questions about fairness are increasingly investigated in the literature. In particular, many authors take a Rawlsian approach to algorithmic fairness. This article aims to identify some complications with this approach: Under which circumstances can Rawls’s original position reasonably be applied to algorithmic fairness decisions? First, it is argued that there are important differences between Rawls’s original position and a (...) parallel algorithmic fairness original position with respect to risk attitudes. Second, it is argued that the application of Rawls’s original position to algorithmic fairness faces a boundary problem in defining relevant stakeholders. Third, it is observed that the definition of the least advantaged, necessary for applying the difference principle, requires some attention in the context of algorithmic fairness. Finally, it is argued that appropriate deliberation in algorithmic fairness contexts often require more knowledge about probabilities than the Rawlsian original position allows. Provided that these complications are duly considered, the thought-experiment of the Rawlsian original position can be useful in algorithmic fairness decisions. (shrink)

The literature on algorithmic fairness has examined exogenous sources of biases such as shortcomings in the data and structural injustices in society. It has also examined internal sources of bias as evidenced by a number of impossibility theorems showing that no algorithm can concurrently satisfy multiple criteria of fairness. This paper contributes to the literature stemming from the impossibility theorems by examining how informational richness affects the accuracy and fairness of predictive algorithms. With the aid of a computer simulation, we (...) show that informational richness is the engine that drives improvements in the performance of a predictive algorithm, in terms of both accuracy and fairness. The centrality of informational richness suggests that classification parity, a popular criterion of algorithmic fairness, should be given relatively little weight. But we caution that the centrality of informational richness should be taken with a grain of salt in light of practical limitations, in particular, the so-called bias-variance trade off. (shrink)

Artificial intelligence tools used in employment decision-making cut across the multiple stages of job advertisements, shortlisting, interviews and hiring, and actual and potential bias can arise in each of these stages. One major challenge is to mitigate AI bias and promote fairness in opaque AI systems. This paper argues that the equal opportunity merit principle is an ethical approach for fair AI employment decision-making. Further, explainable AI can mitigate the opacity problem by placing greater emphasis on enhancing the understanding of (...) reasonable users and affected persons as to the AI output. Both the equal opportunity merit principle and explainable AI should be integrated in the design and implementation of AI employment decision-making systems so as to ensure, as far as possible, that the AI output is arrived at through a fair process. (shrink)

In various official documents, the European Union has declared its goal to pursue a citizen-centric governance of digital transformation. Through a critical review of several of these documents, here we show how “citizen-centric” is more a glamouring than a driving concept. De facto, the EU is enabling a federated data system that is corporate-driven, economic-oriented, and GDPR-compliant; in other words, a Digital Single Market (DSM). This leaves out societal and collective-level dimensions of digital transformation—such as social inclusion, digital sovereignty, and (...) environmental sustainability—which are acknowledged, but not operationalized, by the EU as pillars of a citizen-centric governance. Hence, the door is open to a complementary approach to the governance of digital transformation. We argue that, while a federated data model can constitute the tech-legal backbone of the emerging DSM, a commoning of data, as an ecosystemic approach that maintains a societal and collective outlook by default, can represent a complement to enact a truly citizen-centric governance. (shrink)

Artificial intelligence systems have been widely applied to various contexts, including high-stake decision processes in healthcare, banking, and judicial systems. Some developed AI models fail to offer a fair output for specific minority groups, sparking comprehensive discussions about AI fairness. We argue that the development of AI systems is marked by a central paradox: the less participation one stakeholder has within the AI system’s life cycle, the more influence they have over the way the system will function. This means that (...) the impact on the fairness of the system is in the hands of those who are less impacted by it. However, most of the existing works ignore how different aspects of AI fairness are dynamically and adaptively affected by different stages of AI system development. To this end, we present a use case to discuss fairness in the development of corporate wellness programs using smart wearables and AI algorithms to analyze data. The four key stakeholders throughout this type of AI system development process are presented. These stakeholders are called service designer, algorithm designer, system deployer, and end-user. We identify three core aspects of AI fairness, namely, contextual fairness, model fairness, and device fairness. We propose a relative contribution of the four stakeholders to the three aspects of fairness. Furthermore, we propose the boundaries and interactions between the four roles, from which we make our conclusion about the possible unfairness in such an AI developing process. (shrink)