The primary aim of this paper is twofold: firstly, to argue that we can enter into relation of trust with robots and AI systems (automata); and secondly, to provide a comprehensive description of the underlying mechanisms responsible for this relation of trust. To achieve these objectives, the paper first undertakes a critical examination of the main arguments opposing the concept of a trust-based relation with automata. Showing that these arguments face significant challenges that render them untenable, it thereby prepares the (...) ground for the subsequent positive analysis, proposing a framework in which these challenges can be addressed. According to this framework trust does not originate from mere reliability, but rather from an empathic relation with automata. This initial empathic relation elevates the automata to the status of what I will term "Otheroids." The paper then explores how this human-Otheroid relationship inherently possesses the seeds for the development of trust. Finally, it examines how these seeds can grow into a basic form of trust with Otheroids through the establishment of a rich history of interaction. (shrink)

In this paper, I explore the concept of corrective trust as a pathway to mitigate biases, and potentially build or restore mutual trust in relationships characterized by power imbalances, particularly within the context of healthcare. Corrective trust takes place when we actively choose to trust others when our initial mistrust or hesitation to trust is due to biases. However, existing accounts of trust as a special form of reliance present challenges to practicing corrective trust. I propose a non-reductive account that (...) construes trust as a basic (i.e. irreducible to more basic concepts such as reliance) but complex (i.e. consists of cognitive, affective, and normative components) attitude, disentangling it from reliance and thus addressing associated challenges. This redefinition enables a deeper understanding of trust and facilitates the application of corrective trust to mitigate biases in healthcare and beyond. (shrink)

As technology improves, artificial systems are increasingly able to behave in human-like ways: holding a conversation; providing information, advice, and support; or taking on the role of therapist, teacher, or counsellor. This enhanced behavioural complexity, we argue, encourages deeper forms of affective engagement on the part of the human user, with the artificial agent helping to stabilise, subdue, prolong, or intensify a person’s emotional condition. Here, we defend a fictionalist account of human/AI interaction, according to which these encounters involve an (...) elaborate practise of imaginative pretence: a make-believe in which the artificial agent is attributed a life of its own. We attend, specifically, to the temporal characteristics of these fictions, and to what we imagine artificial agents are doing when we are not looking at them. (shrink)

An intuitive conception of objectivity involves an ideal of neutrality—if we’re to engage in objective inquiry, we must try to sideline our prejudices, values, and politics, lest these factors taint inquiry and unduly influence our results. This intuition underlies various “purist” epistemological frameworks, which grant epistemic significance only to “epistemic factors” like evidence or the truth of a belief. Feminist epistemologists typically condemn purist frameworks as inimical to feminist aims. They argue that purist epistemology is divorced from the ineliminably social (...) and value-laden nature of our epistemic practices, and, moreover, that it abstracts away precisely those features that are needed to explain how power shapes knowledge. As they see it, purist frameworks simply lack the conceptual space needed to ask the right kinds of questions. Thus, they go in for “impurist” frameworks, which grant epistemic significance to “non-epistemic” factors, and so shed the confining constraints of purism. I turn the tables on the impurists—where they have connected purist norms with perniciously disengaged epistemology, I connect impurist norms with a creeping anxiety that is diminishing feminist epistemology’s political possibilities. First, I demonstrate that it’s increasingly common for impurist epistemologists to use moral considerations as a means to bypass the evidence, which betrays a lack of faith that the evidence will bear out feminist commitments. Second, I demonstrate that there is needless pessimism about our capacity to understand, as evidenced by the increasingly widespread idea that privileged social positions place robust limits on what people can know about oppression that they don’t personally experience. In both cases, moral and political considerations are invoked to give the illusion of being socially engaged, while in fact they allow people to give up on the difficult project of figuring things out. An extended case study on rape culture illustrates the stakes of doing inquiry that is only superficially socially engaged; I show that the two anxious traps I’ve identified have prevented contemporary feminist theorists from detecting and rooting out racist distortion in their accounts of rape culture because both traps interfere with a genuinely intersectional approach to feminist theorizing. (shrink)

In this open access book, Timothy Aylsworth and Clinton Castro draw on the deep well of Kantian ethics to argue that we have moral duties, both to ourselves and to others, to protect our autonomy from the threat posed by the problematic use of technology. The problematic use of technologies like smartphones threatens our autonomy in a variety of ways, and critics have only begun to appreciate the vast scope of this problem. In the last decade, we have seen a (...) flurry of books making “self-help” arguments about how we could live happier, more fulfilling lives if we were less addicted to our phones. But none of these authors see this issue as one involving a moral duty to protect our autonomy. (shrink)

In this chapter, we turn our attention to the effects of the attention economy on our ability to act autonomously as a group. We begin by clarifying which sorts of groups we are concerned with, which are structured groups (groups sufficiently organized that it makes sense to attribute agency to the group itself). Drawing on recent work by Purves and Davis (2022), we describe the essential roles of trust (i.e., depending on groups to fulfill their commitments) and trustworthiness (i.e., the (...) property of a group that makes trusting them fitting) in autonomous group action, with particular emphasis on democratic institutions (which we view as group agents) and democratic legitimacy (which depends on trust and trustworthiness). We then explain how engagement maximization promotes polarization, which is detrimental to trust and trustworthiness and, in turn, democratic legitimacy and democratic institutions. We close by considering what groups might do to protect themselves from the threat posed to them by the attention economy. (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)

Theories of rational decision making often abstract away from computational and other resource limitations faced by real agents. An alternative approach known as resource rationality puts such matters front and center, grounding choice and decision in the rational use of finite resources. Anticipated by earlier work in economics and in computer science, this approach has recently seen rapid development and application in the cognitive sciences. Here, the theory of rationality plays a dual role, both as a framework for normative assessment (...) and as a source of scientific hypotheses about how mental processes in fact work. The latter project, often called rational analysis, depends for its success on a fine-grained characterization of the computational problem facing a decision maker, which may in turn depend on realistic assumptions about what the relevant agent is like. As a consequence, resource rationality involves a delicate, but often fruitful interplay between the normative and the descriptive. (shrink)

According to the thesis of doxastic wronging, our beliefs can non-derivatively wrong others. A recent criticism of this view claims that proponents of the doxastic wronging thesis have no principled grounds for denying that credences can likewise non-derivatively wrong, so they must countenance pervasive conflicts between morality and epistemic rationality. This paper defends the thesis of doxastic wronging from this objection by arguing that belief bears distinctive relationships to inquiry and blame that can explain why beliefs, but not credences, can (...) non-derivatively wrong. First, forming a belief (but not updating one’s credence) closes inquiry, and suspending judgment (but not updating one’s credence) opens inquiry. Consequently, beliefs can distinctively wrong others by prematurely closing inquiry or inappropriately opening inquiry. Second, beliefs (but not credences) can constitute blame. Unfittingly blaming someone can wrong them, and hence beliefs which constitute unfitting blame can distinctively wrong. In addition to defending the claim that only beliefs can non-derivatively wrong, this paper gestures towards an ethics of belief which attends to the relationship between belief and attitudes such as inquiry, faith, trust, and blame. (shrink)

Kierkegaard's pseudonym Johannes Climacus famously defines faith as a risky “venture” that requires “holding fast” to “objective uncertainty.” Yet puzzlingly, he emphasizes that faith requires resolute conviction and certainty. Moreover, Climacus claims that all beliefs about contingent propositions about the external world “exclude doubt” and “nullify uncertainty,” but also that uncertainty is “continually present” in these very same beliefs. This paper argues that these apparent contradictions can be resolved by interpreting Climacus as a belief‐credence dualist. That is, Climacus holds that (...) beliefs and credences (i.e., degrees of belief) are two irreducibly distinct types of mental states. Beliefs are resolutions that close inquiry, thereby excluding doubt and reflecting subjective certainty by disregarding the possibility of error. Credences, by contrast, reflect assessments of evidential probabilities, thereby encoding a recognition of “objective uncertainty” by acknowledging the possibility of error. In addition to solving a vexing interpretive puzzle and showing how Kierkegaard anticipates contemporary views about the nature of belief and credence, this paper demonstrates that Kierkegaard developed a sophisticated account of the nature of belief, doubt, and certainty that merits serious philosophical consideration. (shrink)

Technologies can have profound effects on social moral systems. Is there any way to systematically investigate and anticipate these potential effects? This paper aims to contribute to this emerging field on inquiry through a case study method. It focuses on two core human values—truth and trust—describes their structural properties and conceptualisations, and then considers various mechanisms through which technology is changing and can change our perspective on those values. In brief, the paper argues that technology is transforming these values by (...) changing the costs/benefits of accessing them; allowing us to substitute those values for other, closely-related ones; increasing their perceived scarcity/abundance; and disrupting traditional value-gatekeepers. This has implications for how we study other, technologically-mediated, value changes. (shrink)

Social media has invaded our private, professional, and public lives. While corporations continue to portray social media as a celebration of self-expression and freedom, public opinion, by contrast, seems to have decidedly turned against social media. Yet we continue to use it just the same. What is social media, and how should we live with it? Is it the promise of a happier and more interconnected humanity, or a vehicle for toxic self-promotion? In this essay I examine the very structure (...) of social media communications in order to sketch how we should engage with social media. Social media communications are, I argue, a public communication of private content. This allows connections to be made with others in ways that would not otherwise be possible; however, it also submits the private to a status competition, which in turn is linked to mental health challenges. A ‘virtuous’ engagement with social media means being aware of these dynamics, and choosing to subordinate social media to other, more important goods. (shrink)

A widespread assumption in debates about trust and trustworthiness is that the evaluative norms of principal interest on the trustor’s side of a cooperative exchange regulate trusting attitudes and performances whereas those on the trustee’s side regulate dispositions to respond to trust. The aim here will be to highlight some unnoticed problems with this asymmetrical picture – and in particular, how it elides certain key evaluative norms on both the trustor’s and trustee’s side the satisfaction of which are critical to (...) successful cooperative exchanges – and to show that a symmetrical, ‘achievement-first’ approach to theorising about trust and trustworthiness (and their relation to each other) has important advantages by comparison. The view I develop is guided by a structural analogy with practical reasoning. Just as practical reasoning is working as it should only when there is realisation (knowledge and action) of states (belief and intention) with reverse directions of fits (mind-to-world and world-to-mind), likewise, cooperation between trustor and trustee is functioning as it should only when there is an analogous kind of realisation on both sides of the cooperative exchange – viz., when the trustor ‘matches’ her achievement in trusting (an achievement in fitting reliance to reciprocity) with the trustee’s achievement in responding to trust (an achievement in fitting reciprocity to reliance). An upshot of viewing cooperation between trustor and trustee as exhibiting achievement-theoretic structure is that we will be better positioned to subsume trustworthiness (and its cognates on the trustee’s side), like trust, under a wider suite of evaluative norms that regulate attempts, dispositions, and achievements symmetrically on both sides of a cooperative exchange, with ‘matching achievements’ as the gold standard. (shrink)

In this paper, I argue for a probabilistic theory of trust, and the plausibility of “trustworthy AI” in which we trust (as opposed to mere reliance). I show that the current trust theories cannot accommodate trust pertaining to AI, and I propose an alternative probabilistic theory, which accounts for the four major types of AI-related trust: an AI agent’s trust in another AI agent, a human agent’s trust in an AI agent, an AI agent’s trust in a human agent, and (...) an AI agent’s trust in an object (including mental and complex objects). I draw a broadly neglected distinction between transitive and intransitive senses of trust, each of which calls for a distinctive semantical theory. Based on this distinction, I classify the current theories into the theories of trust and theories of trustworthiness, showing that the current theories fail to model some of the major types of AI-related trust; while the proposed conditional probabilistic theory of trust and theory of trustworthiness, unlike the current trust theories, is scalable, and they would also accommodate major types of trust in non-AI, including interpersonal trust, reciprocal trust, one-sided trust, as well as trust in objects—e.g., thoughts, theories, data, algorithms, systems, and institutions. (shrink)

It is well-known that racism is encoded into the social practices and institutions of medicine. Less well-known is that racism is encoded into the material artifacts of medicine. We argue that many medical devices are not merely biased, but materialize oppression. An oppressive device exhibits a harmful bias that reflects and perpetuates unjust power relations. Using pulse oximeters and spirometers as case studies, we show how medical devices can materialize oppression along various axes of social difference, including race, gender, class, (...) and ability. Our account uses political philosophy and cognitive science to give a theoretical basis for understanding materialized oppression, explaining how artifacts encode and carry oppressive ideas from the past to the present and future. Oppressive medical devices present a moral aggregation problem. To remedy this problem, we suggest redundantly layered solutions that are coordinated to disrupt reciprocal causal connections between the attitudes, practices, and artifacts of oppressive systems. (shrink)

In this paper, I distinguish three general approaches to public trust in science, which I call the individual approach, the semi-social approach, and the social approach, and critically examine their proposed solutions to what I call the problem of harmful distrust. I argue that, despite their differences, the individual and the semi-social approaches see the solution to the problem of harmful distrust as consisting primarily in trying to persuade individual citizens to trust science and that both approaches face two general (...) problems, which I call the problem of overidealizing science and the problem of overburdening citizens. I then argue that in order to avoid these problems we need to embrace a (thoroughly) social approach to public trust in science, which emphasizes the social dimensions of the reception, transmission, and uptake of scientific knowledge in society and the ways in which social forces influence both positively and negatively the trustworthiness of science. (shrink)

In her BBC Reith Lectures on Trust, Onora O’Neill offers a short, but biting, criticism of transparency. People think that trust and transparency go together but in reality, says O'Neill, they are deeply opposed. Transparency forces people to conceal their actual reasons for action and invent different ones for public consumption. Transparency forces deception. I work out the details of her argument and worsen her conclusion. I focus on public transparency – that is, transparency to the public over expert domains. (...) I offer two versions of the criticism. First, the epistemic intrusion argument: The drive to transparency forces experts to explain their reasoning to non-experts. But expert reasons are, by their nature, often inaccessible to non-experts. So the demand for transparency can pressure experts to act only in those ways for which they can offer public justification. Second, the intimate reasons argument: In many cases of practical deliberation, the relevant reasons are intimate to a community and not easily explicable to those who lack a particular shared background. The demand for transparency, then, pressures community members to abandon the special understanding and sensitivity that arises from their particular experiences. Transparency, it turns out, is a form of surveillance. By forcing reasoning into the explicit and public sphere, transparency roots out corruption — but it also inhibits the full application of expert skill, sensitivity, and subtle shared understandings. The difficulty here arises from the basic fact that human knowledge vastly outstrips any individual’s capacities. We all depend on experts, which makes us vulnerable to their biases and corruption. But if we try to wholly secure our trust — if we leash groups of experts to pursuing only the goals and taking only the actions that can be justified to the non-expert public — then we will undermine their expertise. We need both trust and transparency, but they are in essential tension. This is a deep practical dilemma; it admits of no neat resolution, but only painful compromise. (shrink)

In this paper, I describe how artifacts and autobiographical memory are integrated into new systemic wholes, allowing us to remember our personal past in a more reliable and detailed manner. After discussing some empirical work on lifelogging technology, I elaborate on the dimension of autobiographical dependency, which is the degree to which we depend on an object to be able to remember a personal experience. When this dependency is strong, we integrate information in the embodied brain and in an object (...) to reconstruct an autobiographical memory. In such cases, autobiographical memory is extended or distributed. (shrink)

Our life with art is suffused with trust. We don’t just trust one another’s aesthetic testimony; we trust one another’s aesthetic actions. Audiences trust artists to have made it worth their while; artists trust audiences to put in the effort. Without trust, audiences would have little reason to put in the effort to understand difficult and unfamiliar art. I offer a theory of aesthetic trust, which highlights the importance of trust in aesthetic sincerity. We trust in another’s aesthetic sincerity when (...) we rely on them to fulfill their commitments to act for aesthetic reasons — rather than for, say, financial, social, or political reasons. We feel most thoroughly betrayed by an artist, not when they make bad art, but when they sell out. This teaches us something about the nature of trust in general. According to many standard theories, trust involves thinking the trusted to be cooperative or good-natured. But trust in aesthetic sincerity is different. We trust artists to be true to their own aesthetic sensibility, which might involve selfishly ignoring their audience’s needs. Why do we care so much about an artist’s sincerity, rather than merely trusting them to make good art? We emphasize sincerity when wish to encourage originality, rather than to demand success along predictable lines. And we ask for sincerity when our goal is to discover a shared sensibility. In moral life, we often try to force convergence through coordinated effort. But in aesthetic life, we often hope for the lovely discovery that our sensibilities were similar all along. And for that we need to ask for sincerity, rather than overt coordination. (shrink)

In analyzing oppressive systems like racism, social theorists have articulated accounts of the dynamic interaction and mutual dependence between psychological components, such as individuals’ patterns of thought and action, and social components, such as formal institutions and informal interactions. We argue for the further inclusion of physical components, such as material artifacts and spatial environments. Drawing on socially situated and ecologically embedded approaches in the cognitive sciences, we argue that physical components of racism are not only shaped by, but also (...) shape psychological and social components of racism. Indeed, while our initial focus is on racism and racist things, we contend that our framework is also applicable to other oppressive systems, including sexism, classism, and ableism. This is because racist things are part of a broader class of oppressive things, which are material artifacts and spatial environments that are in congruence with an oppressive system. (shrink)

In this paper, I describe how artifacts and autobiographical memory are integrated into new systemic wholes, allowing us to remember our personal past in a more reliable and detailed manner. After discussing some empirical work on lifelogging technology, I elaborate on the dimension of autobiographical dependency, which is the degree to which we depend on an object to be able to remember a personal experience. When this dependency is strong, we integrate information in the embodied brain and in an object (...) to reconstruct an autobiographical memory. In such cases, autobiographical memory is extended or distributed. (shrink)

In the social epistemology of scientific knowledge, it is largely accepted that relationships of trust, not just reliance, are necessary in contemporary collaborative science characterised by relationships of opaque epistemic dependence. Such relationships of trust are taken to be possible only between agents who can be held accountable for their actions. But today, knowledge production in many fields makes use of AI applications that are epistemically opaque in an essential manner. This creates a problem for the social epistemology of scientific (...) knowledge, as scientists are now epistemically dependent on AI applications that are not agents, and therefore not appropriate candidates for trust. (shrink)

The doxastic view (DV) of trust maintains that trust essentially involves belief. In a recent paper, Arnon Keren (Citation2020) gestures toward a new objection to the view, labeled Trust’s Meno Problem (TMP), which calls into question the DV’s ability to explain the widely held intuition that trust has distinct and indispensable value. As of yet, there has been no attempt to take up TMP on behalf of DV. This paper aims to fill precisely this lacuna. I do so in three (...) main stages. In §1 I contextualize and elucidate the problem, to which Keren gestures but does not address in detail. In §2 I disambiguate multiple possible interpretations of TMP, seeking to identify the most philosophically challenging. Finally, in §3, I argue that DV can solve even this interpretation. In order to do so, I make use ofthe highly plausible claim we find in the work of Katherine Hawley (Citation2012, 2019): that trust pays a compliment to the trustee. The payoffs of exploring the doxastic view in the context of Trust’s Meno Problem are twofold: we better understand the nature of the problem itself, and we see that the doxastic view can give a satisfying answer. (shrink)

It is often taken for granted that monitoring stands in some kind of tension with trusting (e.g., Hieronymi 2008; Wanderer and Townsend 2013; Nguyen forthcoming; McMyler 2011, Castelfranchi and Falcone 2000; Frey 1993; Dasgupta 1988, Litzky et al. 2006) — especially three-place trust (i.e., A trusts B to X), but sometimes also two-place trust (i.e., A trusts B, see, e.g., Baier 1986). Using a case study involving relationship breakdown, repair, and formation, I will argue there are some ways in which (...) monitoring can be conducive to two-place trust, and to instances of three-place trust that are likely to be repeated over time—especially when previously established two-place trust has broken down. The result, I hope, is not any kind of abandoning of the important idea that monitoring can undermine trust, but an appreciation of where the conflict between monitoring and trust doesn’t lie – one from which future work will hopefully be better positioned to illuminate where exactly the conflict is. (shrink)

Much of what we know we know through testimony, and knowing on the basis of testimony requires some degree of trust in speakers. Trust is therefore very valuable. But in trusting, we expose ourselves to risks of harm and betrayal. It is therefore important to trust well. In this paper, I discuss two recent cases of the betrayal of trust in (broadly) academic contexts: one involving hoax submissions to journals, the other faking an identity on social media. I consider whether (...) these betrayals suggest that we ought to be less trusting in contexts like these. I argue that we should not: the acquisition of knowledge is dependent on trust, and we cannot intentionally reduce the extent to which we trust in these kinds of contexts without risking destroying it utterly. Instead, we must trust in our epistemic networks and the way they work to filter out deception. (shrink)

This paper argues that the widespread belief that interactions between blockchains and their users are trust-free is inaccurate and misleading, since this belief not only overlooks the vital role played by trust in the lack of knowledge and control but also conceals the moral and normative relevance of relying on blockchain applications. The paper reaches this argument by providing a close philosophical examination of the concept referred to as trust in blockchain technology, clarifying the trustor group, the structure, and the (...) normatively loaded nature of this trust relation. The paper ends by critically reflecting on two of the most promising values (decentralization and transparency) that can invite users’ trust in blockchain technology, arguing that there is a tension between the pressing values that are intended to be achieved by developers and the predicament situations caused by current blockchain implementations. (shrink)

The ongoing Post Office scandal in the UK and the 2021 Child Daycare Benefit Scandal in the Netherlands make up exemplary cases of how digital automation has changed and in fact severely harmed trust relations ranging from trust in oneself over trust in social roles, trust in institutions, trust in technology and general trust. By looking closer at how digital automation in these cases generated ruptures in the lives of ordinary citizens and also affected the involved institutions and society at (...) large I discuss how we may begin to discuss the impact of digitalization on trust in societies. I start with an introduction to the panoramic view of trust, followed by a short description of the two scandals. Based on the exploration of the two cases I take an ex post facto perspective on the dynamic social events as they unfolded in the scandals in order to analyze how the introduction of digital automation tools affected trust relations. I draw attention to some of the most noticeable changes in the four relations of trust, which is trust in institutions, interpersonal trust, trust in oneself and trust in technology. (shrink)

Should we be worried that the concept of trust is increasingly used when we assess non-human agents and artefacts, say robots and AI systems? Whilst some authors have developed explanations of the concept of trust with a view to accounting for trust in AI systems and other non-agents, others have rejected the idea that we should extend trust in this way. The article advances this debate by bringing insights from conceptual engineering to bear on this issue. After setting up a (...) target concept of trust in terms of four functional desiderata (trust-reliance distinction, explanatory strength, tracking affective responses, and accounting for distrust), I analyze how agential vs. non-agential accounts can satisfy these. A final section investigates how ‘non-ideal’ circumstances—that is, circumstances where the manifest and operative concept use diverge amongst concept users—affect our choice about which rendering of trust is to be preferred. I suggest that some prominent arguments against extending the language of trust to non-agents are not decisive and reflect on an important oversight in the current debate, namely a failure to address how narrower, agent-centred accounts curtail our ability to distrust non-agents. (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)

According to widely accepted views in metasemantics, the outputs of chatbots and other artificial text generators should be meaningless. They aren’t produced with communicative intentions and the systems producing them are not following linguistic conventions. Nevertheless, chatbots have assumed roles in customer service and healthcare, they are spreading information and disinformation and, in some cases, it may be more rational to trust the outputs of bots than those of our fellow human beings. To account for the epistemic role of chatbots (...) in our society, we need to reconcile these observations. This paper argues that our engagement with chatbots should be understood as a form of prop-oriented make-believe; the outputs of chatbots are literally meaningless but fictionally meaningful. With the make-believe approach, we can understand how chatbots can provide us with knowledge of the world through quasi-testimony while preserving our metasemantic theories. This account also helps to connect the study of chatbots with the epistemology of scientific instruments. (shrink)

It is easier to talk frankly to a person whom one trusts. It is also easier to agree with a scientist whom one trusts. Even though in both cases the psychological state that underlies the behavior is called ‘trust’, it is controversial whether it is a token of the same psychological type. Trust can serve an affective, epistemic, or other social function, and comes to interact with other psychological states in a variety of ways. The way that the functional role (...) of trust changes across contexts and objects is further complicated when communities and individuals mediate it through technologies, and even more so when that mediation involves artificial intelligence (AI) and machine learning (ML). In this chapter I look at the ways in which trust in institutions, and specifically the medical profession, is affected by the use of AI and ML. There are two key elements of this analysis. The first is a disanalogy between institutional trust in medicine and institutional trust in science (Irzik and Kurtulmus 2021, 2019; Kitcher 2001). I note that as AI and ML become a more prominent part of medicine, trust in a medical institution becomes more like trust in a scientific institution. This is problematic for institutional trust in medicine and the practice of medicine, since institutional trust in science has been undermined by, among other things, the spread of misinformation online and the replication crisis (Romero 2019). There is also a strong analogy between the psychological state of the person who trusts a scientific report or testimony and the psychological state of a patient who trusts individual recommendations made by a medical professional in a clinical setting. In both cases, institutional trust makes it less likely that a mistake or malfeasance will result in reactive attitudes, such as blame or anger, directed at other individual members of that institution. However, it also renders people vulnerable enough to blame the institution itself. This, with time, can erode trust in the institution and naturally leads to policy recommendations that aim to preserve institutional trust. I survey two ways in which that can be done with institutional trust in medicine in the age of AI and ML. (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)