Computational methods such as computer simulations, Monte Carlo methods, and agent-based modeling have become the dominant techniques in many areas of science. Extending Ourselves contains the first systematic philosophical account of these new methods, and how they require a different approach to scientific method. Paul Humphreys draws a parallel between the ways in which such computational methods have enhanced our abilities to mathematically model the world, and the more familiar ways in which scientific instruments have expanded our access to the (...) empirical world. This expansion forms the basis for a new kind of empiricism, better suited to the needs of science than the older anthropocentric forms of empiricism. Human abilities are no longer the ultimate standard of epistemological correctness.Humphreys also includes arguments for the primacy of properties rather than objects, the need to consider technological constraints when appraising scientific methods, and a detailed account of how the path from computational template to scientific application is constructed. This last feature allows us to hold a form of selective realism in which anti-realist arguments based on formal reconstructions of theories can be avoided. One important consequence of the rise of computational methods is that the traditional organization of the sciences is being replaced by an organization founded on computational templates.Extending Ourselves will be of interest to philosophers of science, epistemologists, and to anyone interested in the role played by computers in modern science. (shrink)

The use of black box algorithms in medicine has raised scholarly concerns due to their opaqueness and lack of trustworthiness. Concerns about potential bias, accountability and responsibility, patient autonomy and compromised trust transpire with black box algorithms. These worries connect epistemic concerns with normative issues. In this paper, we outline that black box algorithms are less problematic for epistemic reasons than many scholars seem to believe. By outlining that more transparency in algorithms is not always necessary, and by explaining that (...) computational processes are indeed methodologically opaque to humans, we argue that the reliability of algorithms provides reasons for trusting the outcomes of medical artificial intelligence. To this end, we explain how computational reliabilism, which does not require transparency and supports the reliability of algorithms, justifies the belief that results of medical AI are to be trusted. We also argue that several ethical concerns remain with black box algorithms, even when the results are trustworthy. Having justified knowledge from reliable indicators is, therefore, necessary but not sufficient for normatively justifying physicians to act. This means that deliberation about the results of reliable algorithms is required to find out what is a desirable action. Thus understood, we argue that such challenges should not dismiss the use of black box algorithms altogether but should inform the way in which these algorithms are designed and implemented. When physicians are trained to acquire the necessary skills and expertise, and collaborate with medical informatics and data scientists, black box algorithms can contribute to improving medical care. (shrink)

Mainstream epistemology is a highly theoretical and abstract enterprise. Traditional epistemologists rarely present their deliberations as critical to the practical problems of life, unless one supposes—as Hume, for example, did not—that skeptical worries should trouble us in our everyday affairs. But some issues in epistemology are both theoretically interesting and practically quite pressing. That holds of the problem to be discussed here: how laypersons should evaluate the testimony of experts and decide which of two or more rival experts is most (...) credible. It is of practical importance because in a complex, highly specialized world people are constantly confronted with situations in which, as comparative novices, they must turn to putative experts for intellectual guidance or assistance. It is of theoretical interest because the appropriate epistemic considerations are far from transparent; and it is not clear how far the problems lead to insurmountable skeptical quandaries. This paper does not argue for flat-out skepticism in this domain; nor, on the other hand, does it purport to resolve all pressures in the direction of skepticism. It is an exploratory paper, which tries to identify problems and examine some possible solutions, not to establish those solutions definitively. (shrink)

Scientists depend on complex computational systems that are often ineliminably opaque, to the detriment of our ability to give scientific explanations and detect artifacts. Some philosophers have s...

The complementary properties and functions of cognitive artifacts and other external resources are integrated into the human cognitive system to varying degrees. The goal of this paper is to develop some of the tools to conceptualize this complementary integration between agents and artifacts. It does so by proposing a multidimensional framework, including the dimensions of information flow, reliability, durability, trust, procedural transparency, informational transparency, individualization, and transformation. The proposed dimensions are all matters of degree and jointly they constitute a multidimensional (...) space in which situated cognitive systems can be located and have certain dimensional configurations. These dimensions provide a new perspective on the conditions for cognitive extension. They are, however, not meant to provide a set of necessary and sufficient conditions, but to provide a toolbox for investigating the degree and nature of the integration of agent and artifact into “new systemic wholes”. The higher a situated system scores on the proposed dimensions, the more functional integration occurs, and the more tightly coupled the system is. (shrink)

Advanced AI systems are rapidly making their way into medical research and practice, and, arguably, it is only a matter of time before they will surpass human practitioners in terms of accuracy, reliability, and knowledge. If this is true, practitioners will have a prima facie epistemic and professional obligation to align their medical verdicts with those of advanced AI systems. However, in light of their complexity, these AI systems will often function as black boxes: the details of their contents, calculations, (...) and procedures cannot be meaningfully understood by human practitioners. When AI systems reach this level of complexity, we can also speak of black-box medicine. In this paper, we want to argue that black-box medicine conflicts with core ideals of patient-centered medicine. In particular, we claim, black-box medicine is not conducive for supporting informed decision-making based on shared information, shared deliberation, and shared mind between practitioner and patient. (shrink)

In recent years, a plethora of high-profile scientific publications has been reporting about machine learning algorithms outperforming clinicians in medical diagnosis or treatment recommendations. This has spiked interest in deploying relevant algorithms with the aim of enhancing decision-making in healthcare. In this paper, we argue that instead of straightforwardly enhancing the decision-making capabilities of clinicians and healthcare institutions, deploying machines learning algorithms entails trade-offs at the epistemic and the normative level. Whereas involving machine learning might improve the accuracy of medical (...) diagnosis, it comes at the expense of opacity when trying to assess the reliability of given diagnosis. Drawing on literature in social epistemology and moral responsibility, we argue that the uncertainty in question potentially undermines the epistemic authority of clinicians. Furthermore, we elucidate potential pitfalls of involving machine learning in healthcare with respect to paternalism, moral responsibility and fairness. At last, we discuss how the deployment of machine learning algorithms might shift the evidentiary norms of medical diagnosis. In this regard, we hope to lay the grounds for further ethical reflection of the opportunities and pitfalls of machine learning for enhancing decision-making in healthcare. (shrink)

Although decision‐making algorithms are not new to medicine, the availability of vast stores of medical data, gains in computing power, and breakthroughs in machine learning are accelerating the pace of their development, expanding the range of questions they can address, and increasing their predictive power. In many cases, however, the most powerful machine learning techniques purchase diagnostic or predictive accuracy at the expense of our ability to access “the knowledge within the machine.” Without an explanation in terms of reasons or (...) a rationale for particular decisions in individual cases, some commentators regard ceding medical decision‐making to black box systems as contravening the profound moral responsibilities of clinicians. I argue, however, that opaque decisions are more common in medicine than critics realize. Moreover, as Aristotle noted over two millennia ago, when our knowledge of causal systems is incomplete and precarious—as it often is in medicine—the ability to explain how results are produced can be less important than the ability to produce such results and empirically verify their accuracy. (shrink)

I outline a number of parallels between trust and distrust, emphasising the significance of situations in which both trust and distrust would be an imposition upon the (dis)trustee. I develop an account of both trust and distrust in terms of commitment, and argue that this enables us to understand the nature of trustworthiness. Note that this article is available open access on the journal website.

If understanding is factive, the propositions that express an understanding are true. I argue that a factive conception of understanding is unduly restrictive. It neither reflects our practices in ascribing understanding nor does justice to contemporary science. For science uses idealizations and models that do not mirror the facts. Strictly speaking, they are false. By appeal to exemplification, I devise a more generous, flexible conception of understanding that accommodates science, reflects our practices, and shows a sufficient but not slavish sensitivity (...) to the facts. (shrink)

Organizations such as the EU High-Level Expert Group on AI and the IEEE have recently formulated ethical principles and (moral) values that should be adhered to in the design and deployment of artificial intelligence (AI). These include respect for autonomy, non-maleficence, fairness, transparency, explainability, and accountability. But how can we ensure and verify that an AI system actually respects these values? To help answer this question, I propose an account for determining when an AI system can be said to embody (...) certain values. This account understands embodied values as the result of design activities intended to embed those values in such systems. AI systems are here understood as a special kind of sociotechnical system that, like traditional sociotechnical systems, are composed of technical artifacts, human agents, and institutions but—in addition—contain artificial agents and certain technical norms that regulate interactions between artificial agents and other elements of the system. The specific challenges and opportunities of embedding values in AI systems are discussed, and some lessons for better embedding values in AI systems are drawn. (shrink)

Deep neural networks have become increasingly successful in applications from biology to cosmology to social science. Trained DNNs, moreover, correspond to models that ideally allow the prediction of new phenomena. Building in part on the literature on ‘eXplainable AI’, I here argue that these models are instrumental in a sense that makes them non-explanatory, and that their automated generation is opaque in a unique way. This combination implies the possibility of an unprecedented gap between discovery and explanation: When unsupervised models (...) are successfully used in exploratory contexts, scientists face a whole new challenge in forming the concepts required for understanding underlying mechanisms. (shrink)

The usefulness of machine learning algorithms has led to their widespread adoption prior to the development of a conceptual framework for making sense of them. One common response to this situation is to say that machine learning suffers from a “black box problem.” That is, machine learning algorithms are “opaque” to human users, failing to be “interpretable” or “explicable” in terms that would render categorization procedures “understandable.” The purpose of this paper is to challenge the widespread agreement about the existence (...) and importance of a black box problem. The first section argues that “interpretability” and cognates lack precise meanings when applied to algorithms. This makes the concepts difficult to use when trying to solve the problems that have motivated the call for interpretability. Furthermore, since there is no adequate account of the concepts themselves, it is not possible to assess whether particular technical features supply formal definitions of those concepts. The second section argues that there are ways of being a responsible user of these algorithms that do not require interpretability. In many cases in which a black box problem is cited, interpretability is a means to a further end such as justification or non-discrimination. Since addressing these problems need not involve something that looks like an “interpretation” of an algorithm, the focus on interpretability artificially constrains the solution space by characterizing one possible solution as the problem itself. Where possible, discussion should be reformulated in terms of the ends of interpretability. (shrink)

This paper offers a sizeable menu of approaches to what it means to be an expert. Is it a matter of reputation within a community, or a matter of what one knows independently of reputation? An initial proposal characterizes expertise in dispositional terms—an ability to help other people get answers to difficult questions or execute difficult tasks. What cognitive states, however, ground these abilities? Do the grounds consist in “veritistic” states or in terms of evidence or justifiedness? To what extent (...) is expertise a matter of superior knowledge or other factors? Some authors seek to debunk the notion of expertise entirely. The present approach resists this stance, but doesn’t dispute the variability and fluidness of the concept. Even more challenging is the problem of how laypersons can determine who is the superior expert, especially when experts disagree. (shrink)

Several philosophical issues in connection with computer simulations rely on the assumption that results of simulations are trustworthy. Examples of these include the debate on the experimental role of computer simulations :483–496, 2009; Morrison in Philos Stud 143:33–57, 2009), the nature of computer data Computer simulations and the changing face of scientific experimentation, Cambridge Scholars Publishing, Barcelona, 2013; Humphreys, in: Durán, Arnold Computer simulations and the changing face of scientific experimentation, Cambridge Scholars Publishing, Barcelona, 2013), and the explanatory power of (...) computer simulations :277–292, 2008; Durán in Int Stud Philos Sci 31:27–45, 2017). The aim of this article is to show that these authors are right in assuming that results of computer simulations are to be trusted when computer simulations are reliable processes. After a short reconstruction of the problem of epistemic opacity, the article elaborates extensively on computational reliabilism, a specified form of process reliabilism with computer simulations located at the center. The article ends with a discussion of four sources for computational reliabilism, namely, verification and validation, robustness analysis for computer simulations, a history of successful implementations, and the role of expert knowledge in simulations. (shrink)

Cognitive science normally takes the individual agent as its unit of analysis. In many human endeavors, however, the outcomes of interest are not determined entirely by the information processing properties of individuals. Nor can they be inferred from the properties of the individual agents, alone, no matter how detailed the knowledge of the properties of those individuals may be. In commercial aviation, for example, the successful completion of a flight is produced by a system that typically includes two or more (...) pilots interacting with each other and with a suite of technological devices. This article presents a theoretical framework that takes a distributed, socio‐technical system rather than an individual mind as its primary unit of analysis. This framework is explicitly cognitive in that it is concerned with how information is represented and how representations are transformed and propagated in the performance of tasks. An analysis of a memory task in the cockpit of a commercial airliner shows how the cognitive properties of such distributed systems can differ radically from the cognitive properties of the individuals who inhabit them. (shrink)

Synthesizing the domains of investigation highlighted in current research in distributed cognition and related fields, this paper offers an initial taxonomy of the overlapping types of resources which typically contribute to distributed or extended cognitive systems. It then outlines a number of key dimensions on which to analyse both the resulting integrated systems and the components which coalesce into more or less tightly coupled interaction over the course of their formation and renegotiation.

This paper tackles the problem of defining what a cognitive expert is. Starting from a shared intuition that the definition of an expert depends upon the conceptual function of expertise, I shed light on two main approaches to the notion of an expert: according to novice-oriented accounts of expertise, experts need to provide laypeople with information they lack in some domain; whereas, according to research-oriented accounts, experts need to contribute to the epistemic progress of their discipline. In this paper, I (...) defend the thesis that cognitive experts should be identified by their ability to perform the latter function rather than the former, as novice-oriented accounts, unlike research-oriented ones, fail to comply with the rules of a functionalist approach to expertise. (shrink)

What is the mission of epistemology, and what is its proper methodology? Such meta-epistemological questions have been prominent in recent years, especially with the emergence of various brands of "naturalistic" epistemology. In this paper, I shall reformulate and expand upon my own meta-epistemological conception (most fully articulated in Goldman, 1986), retaining many of its former ingredients while reconfiguring others. The discussion is by no means confined, though, to the meta-epistemological level. New substantive proposals will also be advanced and defended.

In this paper I argue that Artificial Intelligence and the many data science methods associated with it, such as machine learning and large language models, are first and foremost epistemic technologies. In order to establish this claim, I first argue that epistemic technologies can be conceptually and practically distinguished from other technologies in virtue of what they are designed for, what they do and how they do it. I then proceed to show that unlike other kinds of technology (_including_ other (...) epistemic technologies) AI can be uniquely positioned as an epistemic technology in that it is primarily designed, developed and deployed to be used in _epistemic contexts_ such as inquiry, it is specifically designed, developed and deployed to manipulate _epistemic content_ such as data, _and_ it is designed, developed and deployed to do so particularly through _epistemic operations_ such as prediction and analysis. As has been shown in recent work in the philosophy and ethics of AI (Alvarado, AI and Ethics, 2022a), understanding AI as an epistemic technology will also have significant implications for important debates regarding our relationship to AI technologies. This paper includes a brief overview of such implications, particularly those pertaining to explainability, opacity, trust and even epistemic harms related to AI technologies. (shrink)

One of the key tenets of Linda Zagzebski’s book " Epistemic Authority" is the Preemption Thesis. It says that, when an agent learns that an epistemic authority believes that p, the rational response for her is to adopt that belief and to replace all of her previous reasons relevant to whether p by the reason that the authority believes that p. I argue that such a “Hobbesian approach” to epistemic authority yields problematic results. This becomes especially virulent when we apply (...) Preemption to cases in which the agent and the authority share their belief, maybe even for the same reasons, or in which both have either a positive or a negative graded doxastic attitude toward a given proposition. As an alternative I propose a “Socratic account”, according to which the authority will not only motivate us to adopt her belief, but also provide us with higher-order reasons for re-assigning our own considerations their proper place in the web of reasons for and against the view in question. (shrink)

According to a recent account of epistemic authority proposed by Linda Zagzebski (2012), it is rational for laypersons to believe on authority when they conscientiously judge that the authority is more likely to form true beliefs and avoid false ones than they are in some domain. Christoph Jäger (2016) has recently raised several objections to her view. By contrast, I argue that both theories fail to adequately capture what epistemic authority is, and I offer an alternative account grounded in the (...) abilities that different kinds of authorities are required to possess. (shrink)

In recent years, virtue epistemology has won the attention of a wide range of philosophers. A developed form of the position has been expounded forcefully by Ernest Sosa and represents the most plausible version of reliabilism to date. Through the person of Alvin Plantinga, virtue epistemology has taken philosophy of religion by storm, evoking objections and defenses in a wide variety of journals and volumes. Historically, virtue epistemology has its roots in the work of Thomas Reid, and the explosion of (...) Reid scholarship in the last few years is perhaps both a cause and an effect of recent interest in the position.In this paper I want to examine the virtues and vices of virtue epistemology. My conclusion will be that the position is correct, when qualified appropriately. The central claim of virtue epistemology is that, Gettier problems aside, knowledge is true belief which results from a cognitive virtue. In section one I will clarify this claim with some brief remarks about the nature of virtues in general, and cognitive virtues in particular. In section two I will consider two objections to the theory of knowledge which results. In section three of the paper I will argue that virtue epistemology can be qualified so as to avoid the objections raised in section two. Finally, I will argue that the amendments which solve the objections of section two also allow us to solve a version of the dreaded generality problem. (shrink)

As machine learning has gradually entered into ever more sectors of public and private life, there has been a growing demand for algorithmic explainability. How can we make the predictions of complex statistical models more intelligible to end users? A subdiscipline of computer science known as interpretable machine learning (IML) has emerged to address this urgent question. Numerous influential methods have been proposed, from local linear approximations to rule lists and counterfactuals. In this article, I highlight three conceptual challenges that (...) are largely overlooked by authors in this area. I argue that the vast majority of IML algorithms are plagued by (1) ambiguity with respect to their true target; (2) a disregard for error rates and severe testing; and (3) an emphasis on product over process. Each point is developed at length, drawing on relevant debates in epistemology and philosophy of science. Examples and counterexamples from IML are considered, demonstrating how failure to acknowledge these problems can result in counterintuitive and potentially misleading explanations. Without greater care for the conceptual foundations of IML, future work in this area is doomed to repeat the same mistakes. (shrink)

Extended and distributed cognition theories argue that human cognitive systems sometimes include non-biological objects. On these views, the physical supervenience base of cognitive systems is thus not the biological brain or even the embodied organism, but an organism-plus-artifacts. In this paper, we provide a novel account of the implications of these views for learning, education, and assessment. We start by conceptualising how we learn to assemble extended cognitive systems by internalising cultural norms and practices. Having a better grip on how (...) extended cognitive systems are assembled, we focus on the question: If our cognition extends, how should we educate and assess such extended cognitive systems? We suggest various ways to minimize possible negative effects of extending one’s cognition and to efficiently find and organise (online) information by adopting a virtue epistemology approach. Educational and assessment implications are foregrounded, particularly in the case of Danish students’ use of the Internet during exams. (shrink)

What are experts? Are there only experts in a subjective sense or are there also experts in an objective sense? And how, if at all, may non-experts recognize experts in an objective sense? In this paper, I approach these important questions by discussing Alvin I. Goldman's thoughts about how to define objective epistemic authority and about how non-experts are able to identify experts. I argue that a multiple epistemic desiderata approach is superior to Goldman's purely veritistic approach.

This article reviews some recent history of epistemology, focusing on ways in which the intellectual virtues have been invoked to solve specific epistemological problems. It gives a sense of the contemporary landscape that has emerged, and clarifies some of the disagreements among those who invoke the virtues in epistemology. Furthermore, it explores some epistemological problems in greater detail. It also defends a particular approach in virtue epistemology by displaying its power in addressing these problems. It pursues the idea that a (...) minimalist, reliabilist notion of the intellectual virtues is useful for constructing an account of knowledge. It sets out to support this on the ground that this approach to intellectual virtue can adequately address three major problems in the theory of knowledge: Humean skepticism, the Gettier problem, and the problem of showing that knowledge is more valuable than mere true belief. (shrink)

We address an open problem in the philosophy of artificial intelligence (AI): how to justify the epistemic attitudes we have towards the trustworthiness of AI systems. The problem is important, as providing reasons to believe that AI systems are worthy of trust is key to appropriately rely on these systems in human-AI interactions. In our approach, we consider the trustworthiness of an AI as a time-relative, composite property of the system with two distinct facets. One is the actual trustworthiness of (...) the AI and the other is the perceived trustworthiness of the system as assessed by its users while interacting with it. We show that credences, namely, beliefs we hold with a degree of confidence, are the appropriate attitude for capturing the facets of the trustworthiness of an AI over time. Then, we introduce a reliabilistic account providing justification to the credences in the trustworthiness of AI, which we derive from Tang’s probabilistic theory of justified credence. Our account stipulates that a credence in the trustworthiness of an AI system is justified if and only if it is caused by an assessment process that tends to result in a high proportion of credences for which the actual and perceived trustworthiness of the AI are calibrated. This approach informs research on the ethics of AI and human-AI interactions by providing actionable recommendations on how to measure the reliability of the process through which users perceive the trustworthiness of the system, investigating its calibration to the actual levels of trustworthiness of the AI as well as users’ appropriate reliance on the system. (shrink)

This paper attempts to provide a remedy to a surprising lacuna in the current discussion in the epistemology of expertise, namely the lack of a theory accounting for the epistemic authority of collective agents. After introducing a service conception of epistemic authority based on Alvin Goldman’s account of a cognitive expert, I argue that this service conception is well suited to account for the epistemic authority of collective bodies on a non-summativist perspective, and I show in detail how the defining (...) requirements of an expert can apply to epistemic groups. (shrink)