With automation of routine decisions coupled with more intricate and complex information architecture operating this automation, concerns are increasing about the trustworthiness of these systems. These concerns are exacerbated by a class of artificial intelligence that uses deep learning, an algorithmic system of deep neural networks, which on the whole remain opaque or hidden from human comprehension. This situation is commonly referred to as the black box problem in AI. Without understanding how AI reaches its conclusions, it is an open (...) question to what extent we can trust these systems. The question of trust becomes more urgent as we delegate more and more decision-making to and increasingly rely on AI to safeguard significant human goods, such as security, healthcare, and safety. Models that “open the black box” by making the non-linear and complex decision process understandable by human observers are promising solutions to the black box problem in AI but are limited, at least in their current state, in their ability to make these processes less opaque to most observers. A philosophical analysis of trust will show why transparency is a necessary condition for trust and eventually for judging AI to be trustworthy. A more fruitful route for establishing trust in AI is to acknowledge that AI is situated within a socio-technical system that mediates trust, and by increasing the trustworthiness of these systems, we thereby increase trust in AI. (shrink)

A lot of attention has recently been devoted to the notion of Trustworthy AI (TAI). However, the very applicability of the notions of trust and trustworthiness to AI systems has been called into question. A purely epistemic account of trust can hardly ground the distinction between trustworthy and merely reliable AI, while it has been argued that insisting on the importance of the trustee’s motivations and goodwill makes the notion of TAI a categorical error. After providing an overview of the (...) debate, we contend that the prevailing views on trust and AI fail to account for the ethically relevant and value-laden aspects of the design and use of AI systems, and we propose an understanding of the notion of TAI that explicitly aims at capturing these aspects. The problems involved in applying trust and trustworthiness to AI systems are overcome by keeping apart trust in AI systems and interpersonal trust. These notions share a conceptual core but should be treated as distinct ones. (shrink)

In this article, I will argue in favor of both the ethical and epistemological utility of explanations in artificial intelligence -based medical technology. I will build on the notion of “explicability” due to Floridi, which considers both the intelligibility and accountability of AI systems to be important for truly delivering AI-powered services that strengthen autonomy, beneficence, and fairness. I maintain that explicable algorithms do, in fact, strengthen these ethical principles in medicine, e.g., in terms of direct patient–physician contact, as well (...) as on a longer-term epistemological level by facilitating scientific progress that is informed through practice. With this article, I will therefore attempt to counter arguments against demands for explicable AI in medicine that are based on a notion of “whatever heals is right.” I will elucidate my elaboration on the positive aspects of explicable AI in medicine as well as by pointing out risks of non-explicable AI. (shrink)

With an increasing use of artificial intelligence (AI) systems, theorists have analyzed and argued for the promotion of trust in AI and trustworthy AI. Critics have objected that AI does not have the characteristics to be an appropriate subject for trust. However, this argumentation is open to counterarguments. Firstly, rejecting trust in AI denies the trust attitudes that some people experience. Secondly, we can trust other non‐human entities, such as animals and institutions, so why can we not trust AI systems? (...) Finally, human–AI trust is criticized based on a conception of human–human trust, which does not recognize the distinctiveness of the human–AI relationship. This article aims to refute these counterarguments based on the genealogical analyses of ‘trust’ and ‘trustworthiness’ of Karen Jones and Thomas Simpson, who show that trust and trustworthiness help to overcome vulnerabilities. This function of trust gives reason to use human–human trust as a standard. For this function, it is important that trustees are responsive to trust. While animals and institutions could be responsive, narrow AI systems are unable to be responsive to trust. Therefore, we should not apply trust to AI and instead direct our trust to those who can be responsive to and held responsible for our trust. (shrink)

Engineers are traditionally regarded as trustworthy professionals who meet exacting standards. In this chapter I begin by explicating our trust relationship towards engineers, arguing that it is a linear but indirect relationship in which engineers “stand behind” the artifacts and technological systems that we rely on directly. The chapter goes on to explain how this relationship has become more complex as engineers have taken on two additional aims: the aim of social engineering to create and steer trust between people, and (...) the aim of creating automated systems that take over human tasks and are meant to invite the trust of those who rely on and interact with them. (shrink)

This chapter explores properties that bind individuals, knowledge, and communities, together. Section ‎‎1 introduces Hardwig’s argument from trust in others’ testimonies as entailing that trust is the glue ‎that binds individuals into communities. Section 2 asks “what grounds trust?” by exploring assessment ‎of collaborators’ explanatory responsiveness, formal indicators such as affiliation and credibility, ‎appreciation of peers’ tacit knowledge, game-theoretical considerations, and the role moral character ‎of peers, social biases, and social values play in grounding trust. Section 3 deals with establishing (...) ‎reliability standards for formation and breaching of trust. Different epistemic considerations and their ‎underpinning of inductive risks are examined through various communication routes within a ‎discipline, between disciplines, and to the public. Section 4 examines whether a collective entity can ‎be trusted over and above trust that is given to its individual members. Section 5 deals with the roles ‎technological artifacts play in distributed research and collective knowledge. It presents the common ‎view in which genuine trust cannot, in principle, be accorded to artifacts, so as an opposite view. We ‎show that what counts as a genuine object of trust is relevant to debates about the boundaries of ‎collective agency and as a criterion for extended cognitive systems.‎. (shrink)

Much has been written about the need for trustworthy artificial intelligence (AI), but the underlying meaning of trust and trustworthiness can vary or be used in confusing ways. It is not always clear whether individuals are speaking of a technology’s trustworthiness, a developer’s trustworthiness, or simply of gaining the trust of users by any means. In sociotechnical circles, trustworthiness is often used as a proxy for ‘the good’, illustrating the moral heights to which technologies and developers ought to aspire, at (...) times with a multitude of diverse requirements; or at other times, no specification at all. In philosophical circles, there is doubt that the concept of trust should be applied at all to technologies rather than their human creators. Nevertheless, people continue to intuitively reason about trust in technologies in their everyday language. This qualitative study employed an empirical ethics methodology to address how developers and users define and construct requirements for trust throughout development and use, through a series of interviews. We found that different accounts of trust (rational, affective, credentialist, norms based, relational) served as the basis for individual granting of trust in technologies and operators. Ultimately, the most significant requirement for user trust and assessment of trustworthiness was the accountability of AI developers for the outputs of AI systems, hinging on the identification of accountable moral agents and perceived value alignment between the user and developer’s interests. (shrink)

In this paper, I will identify two problems of trust in an AI-relevant context: a theoretical problem and a practical one. I will identify and address a number of skeptical challenges to an AI-relevant theory of trust. In addition, I will identify what I shall term the ‘scope challenge’, which I take to hold for any AI-relevant theory of trust that purports to be representationally adequate to the multifarious forms of trust and AI. Thereafter, I will suggest how trust-engineering, a (...) position that is intermediate between the modified pure rational-choice account and an account that gives rise to trustworthy AI, might allow us to address the practical problem of trust, before identifying and critically evaluating two candidate trust-engineering approaches. (shrink)

Achieving trustworthy AI is increasingly considered an essential desideratum to integrate AI systems into sensitive societal fields, such as criminal justice, finance, medicine, and healthcare, among others. For this reason, it is important to spell out clearly its characteristics, merits, and shortcomings. This article is the first survey in the specialized literature that maps out the philosophical landscape surrounding trust and trustworthiness in AI. To achieve our goals, we proceed as follows. We start by discussing philosophical positions on trust and (...) trustworthiness, focusing on interpersonal accounts of trust. This allows us to explain why trust, in its most general terms, is to be understood as reliance plus some “extra factor”. We then turn to the first part of the definition provided, i.e., reliance, and analyze two opposing approaches to establishing AI systems’ reliability. On the one hand, we consider transparency and, on the other, computational reliabilism. Subsequently, we focus on debates revolving around the “extra factor”. To this end, we consider viewpoints that most actively resist the possibility and desirability of trusting AI systems before turning to the analysis of the most prominent advocates of it. Finally, we take up the main conclusions of the previous sections and briefly point at issues that remain open and need further attention. (shrink)

Complex technology has become an essential aspect of everyday life. We rely on technology as part of basic infrastructure and repeatedly for tasks throughout the day. Yet, in many cases the relation surpasses mere reliance and evolves to trust in technology. A new, disruptive technology is blockchain. It claims to introduce trustless relationships among its users, aiming to eliminate the need for trust altogether—even being described as “the trust machine”. This paper presents a proposal to adjust the concept of trust (...) in blockchain applications with the tools provided by conceptual engineering. Its goal is to propose a concept of trust, that offers more than a halo term, devoid of the normative depth that the original, philosophical term actually carries. To evaluate possible concepts of trust in blockchain applications, five criteria are proposed: These criteria ensure that the conceptual relation indeed embodies trust, thereby being richer than mere reliance, and that the chosen concept highlights the active role of the trustor. While the concepts of trust in engineers and trust in algorithms have to be discarded, institutional trust emerges as a viable candidate, given some refinements. Ultimately, the concept of trust by normative expectations and attribution is suggested to capture the essence of trust in blockchain applications. (shrink)

It is frequently demanded that AI-based Decision Support Tools (AI-DSTs) ought to be both explainable to, and trusted by, those who use them. The joint pursuit of these two principles is ordinarily believed to be uncontroversial. In fact, a common view is that AI systems should be made explainable so that they can be trusted, and in turn, accepted by decision-makers. However, the moral scope of these two principles extends far beyond this particular instrumental connection. This paper argues that if (...) we were to account for the rich and diverse moral reasons that ground the call for explainable AI, and fully consider what it means to “trust” AI in a descriptively rich sense of the term, we would uncover a deep and persistent tension between the two principles. For explainable AI to usefully serve the pursuit of normatively desirable goals, decision-makers must carefully monitor and critically reflect on the content of an AI-DST’s explanation. This entails a deliberative attitude. Conversely, calls for trust in AI-DSTs imply the disposition to put questions about their reliability out of mind. This entails an unquestioning attitude. As such, the joint pursuit of explainable and trusted AI calls on decision-makers to simultaneously adopt incompatible attitudes towards their AI-DST, which leads to an intractable implementation gap. We analyze this gap and explore its broader implications: suggesting that we may need alternate theoretical conceptualizations of what explainability and trust entail, and/or alternate decision-making arrangements that separate the requirements for trust and deliberation to different parties. (shrink)

The present paper aims at contributing to the ethical debate on the impacts of artificial intelligence (AI) systems on human autonomy. More specifically, it intends to offer a clearer understanding of the design challenges to the effort of aligning driving automation technologies to this ethical value. After introducing the discussion on the ambiguous impacts that AI systems exert on human autonomy, the analysis zooms in on how the problem has been discussed in the literature on connected and automated vehicles (CAVs). (...) On this basis, it is claimed that the issue has been mainly tackled on a fairly general level, and mostly with reference to the controversial issue of crash-optimization algorithms, so that only limited design insights have been drawn. However, integrating ethical analysis and design practices is critical to pursue the implementation of such an important ethical value into CAV technologies. To this aim, it is argued, a more applied approach targeted at examining the impacts on human autonomy of current CAV functions should also be explored. As an example of the intricacy of this task, the case of automated route planning is discussed in some detail. (shrink)

Following the publication of numerous ethical principles and guidelines, the concept of 'Trustworthy AI' has become widely used. However, several AI ethicists argue against using this concept, often backing their arguments with decades of conceptual analyses made by scholars who studied the concept of trust. In this paper, I describe the historical-philosophical roots of their objection and the premise that trust entails a human quality that technologies lack. Then, I review existing criticisms about 'Trustworthy AI' and the consequence of ignoring (...) these criticisms: if the concept of 'Trustworthy AI' is kept being used, we risk attributing responsibilities to agents who cannot be held responsible, and consequently, deteriorate social structures which regard accountability and liability. Nevertheless, despite suggestions to shift the paradigm from 'Trustworthy AI' to 'Reliable AI', I argue that, realistically, this concept will be kept being used. I end by arguing that, ultimately, AI ethics is also about power, social justice, and scholarly activism. Therefore, I propose that community-driven and social justice-oriented ethicists of AI and trust scholars further focus on (a) democratic aspects of trust formation; and (b) draw attention to critical social aspects highlighted by phenomena of distrust. This way, it will be possible to further reveal shifts in power relations, challenge unfair status quos, and suggest meaningful ways to keep the interests of citizens. (shrink)

Many recent AI policies have been structured under labels that follow a particular trend: national or international guidelines, policies or regulations, such as the EU’s and USA’s ‘Trustworthy AI’ and China’s and India’s adoption of ‘Responsible AI’, use a label that follows the recipe of [agentially loaded notion + ‘AI’]. A result of this branding, even if implicit, is to encourage the application by laypeople of these agentially loaded notions to the AI technologies themselves. Yet, these notions are appropriate only (...) when applied to agents, which current AI technologies are not; and the concern is that this misapplication creates an incentive to inappropriately attribute trustworthiness or responsibility to AI technologies. We endeavour to show that we have good reason to avoid any general AI policy that uses agentially loaded labelling. We suggest labelling these policies not in terms of some qualification of AI, but rather in terms of our approach to the technology and its wider development and use context – focusing on being trustworthy and responsible about AI, rather than on trustworthy or responsible AI. (shrink)

In this article, we explore questions about the culture of trustworthy artificial intelligence (AI) through the lens of ecosystems. We draw on the European Commission’s Guidelines for Trustworthy AI and its philosophical underpinnings. Based on the latter, the trustworthiness of an AI ecosystem can be conceived of as being grounded by both the so-called rational-choice and motivation-attributing accounts—i.e., trusting is rational because solution providers deliver expected services reliably, while trust also involves resigning control by attributing one’s motivation, and hence, goals, (...) onto another entity. Our research question is: What aspects contribute to a responsible AI ecosystem that can promote justifiable trustworthiness in a healthcare environment? We argue that especially within devising governance and support aspects of a medical AI ecosystem, considering the so-called motivation-attributing account of trust provides fruitful pointers. There can and should be specific ways and governance structures supporting and nurturing trustworthiness beyond mere reliability. After compiling a list of preliminary requirements for this, we describe the emergence of one particular medical AI ecosystem and assess its compliance with and future ways of improving its functioning as a responsible AI ecosystem that promotes trustworthiness. (shrink)

This paper discusses the epistemology of the Internet of Things [IoT] by focusing on the topic of trust. It presents various frameworks of trust, and argues that the ethical framework of trust is what constitutes our responsibility to reveal desired norms and standards and embed them in other frameworks of trust. The first section briefly presents the IoT and scrutinizes the scarce philosophical work that has been done on this subject so far. The second section suggests that the field of (...) epistemology is not sufficiently capable of dealing with technologies, and presents a possible solution to this problem. It is argued that knowledge is not only social phenomena, but also a technological one, and that in order to address epistemological issues in technology, we need to carefully depart from traditional epistemic analysis and form a new approach that is technological (termed here Techno-Epistemology). The third and fourth sections engage in an epistemic analysis of trust by dividing it in to various frameworks. The last section argues that these various frameworks of trust can be understood to form a trustworthy large-scale socio-technological system, emphasizing the place of ethical trust as constituting our commitment to give proper accounts for all of the other frameworks. (shrink)

The philosophy of trust is a relatively small subfield. Nonetheless, it contains within it many important insights. Our contention in this paper is that careful study of this subfield can b...

El discurso sobre la seguridad que viene planteándose en el ámbito de la Inteligencia Artificial (IA) se caracteriza por un predominio de las perspectivas técnica y normativa. Los efectos ambivalentes de esta tecnología y su consideración como un sistema sociotécnico plantean la necesidad de complementar este discurso integrando una perspectiva moral. Así pues, el objetivo principal de este trabajo consiste en argumentar que la seguridad moral es un elemento indispensable para alcanzar una IA confiable, en virtud de las diversas situaciones (...) donde la dignidad humana puede verse comprometida. (shrink)