As systems based on opaque Artificial Intelligence (AI) continue to flourish in diverse real-world applications, understanding these black box models has become paramount. In response, Explainable AI (XAI) has emerged as a field of research with practical and ethical benefits across various domains. This paper not only highlights the advancements in XAI and its application in real-world scenarios but also addresses the ongoing challenges within XAI, emphasizing the need for broader perspectives and collaborative efforts. We bring together experts from diverse (...) fields to identify open problems, striving to synchronize research agendas and accelerate XAI in practical applications. By fostering collaborative discussion and interdisciplinary cooperation, we aim to propel XAI forward, contributing to its continued success. Our goal is to put forward a comprehensive proposal for advancing XAI. To achieve this goal, we present a manifesto of 27 open problems categorized into nine categories. These challenges encapsulate the complexities and nuances of XAI and offer a road map for future research. For each problem, we provide promising research directions in the hope of harnessing the collective intelligence of interested stakeholders. (shrink)

In a recent paper, Erasmus et al. (2021) defend the idea that the ambiguity of the term “explanation” in explainable AI (XAI) can be solved by adopting any of four different extant accounts of explanation in the philosophy of science: the Deductive Nomological, Inductive Statistical, Causal Mechanical, and New Mechanist models. In this chapter, I show that the authors’ claim that these accounts can be applied to deep neural networks as they would to any natural phenomenon is mistaken. I also (...) provide a more general argument as to why the notion of explainability as it is currently used in the XAI literature bears little resemblance to the traditional concept of scientific explanation. It would be more fruitful to use the label “understandable AI” to avoid the confusion that surrounds the goal and purposes of XAI. In the second half of the chapter, I argue for a pragmatic conception of understanding that is better suited to play the central role attributed to explanation in XAI. Following De Regt (2017) and Kuorikoski and Ylikoski (2015), the conditions of satisfaction for understanding an ML system are fleshed out in terms of an agent’s success in using the system. (shrink)

Although standard Machine Learning models are optimized for making predictions about observations, more and more they are used for making predictions about the results of actions. An important goal of Explainable Artificial Intelligence (XAI) is to compensate for this mismatch by offering explanations about the predictions of an ML-model which ensure that they are reliably action-guiding. As action-guiding explanations are causal explanations, the literature on this topic is starting to embrace insights from the literature on causal models. Here I take (...) a step further down this path by formally defining the causal notions of sufficient explanations and counterfactual explanations. I show how these notions relate to (and improve upon) existing work, and motivate their adequacy by illustrating how different explanations are action-guiding under different circum- stances. Moreover, this work is the first to offer a formal definition of actual causation that is founded entirely in action-guiding explanations. Although the definitions are motivated by a focus on XAI, the analysis of causal explanation and actual causation applies in general. I also touch upon the significance of this work for fairness in AI by showing how actual causation can be used to improve the idea of path-specific counterfactual fairness. (shrink)

Despite the growing popularity of machine learning models in the cyber-security applications ), most of these models are perceived as a black-box. The eXplainable Artificial Intelligence has become increasingly important to interpret the machine learning models to enhance trust management by allowing human experts to understand the underlying data evidence and causal reasoning. According to IDS, the critical role of trust management is to understand the impact of the malicious data to detect any intrusion in the system. The previous studies (...) focused more on the accuracy of the various classification algorithms for trust in IDS. They do not often provide insights into their behavior and reasoning provided by the sophisticated algorithm. Therefore, in this paper, we have addressed XAI concept to enhance trust management by exploring the decision tree model in the area of IDS. We use simple decision tree algorithms that can be easily read and even resemble a human approach to decision-making by splitting the choice into many small subchoices for IDS. We experimented with this approach by extracting rules in a widely used KDD benchmark dataset. We also compared the accuracy of the decision tree approach with the other state-of-the-art algorithms. (shrink)

Artificial intelligent (AI) systems that perform image classification tasks are being used to great success in many application contexts. However, many of these systems are opaque, even to experts. This lack of understanding can be problematic for ethical, legal, or practical reasons. The research field Explainable AI (XAI) has therefore developed several approaches to explain image classifiers. The hope is to bring about understanding, e.g., regarding why certain images are classified as belonging to a particular target class. Most of these (...) approaches use visual explanations. Drawing on Elgin’s work (True enough. MIT Press, Cambridge, 2017), I argue that analyzing what those explanations exemplify can help to assess their suitability for producing understanding. More specifically, I suggest to distinguish between two forms of examples according to their suitability for producing understanding. I call these forms samples and exemplars, respectively. Samples are prone to misinterpretation and thus carry the risk of leading to misunderstanding. Exemplars, by contrast, are intentionally designed or chosen to meet contextual requirements and to mitigate the risk of misinterpretation. They are thus preferable for bringing about understanding. By reviewing several XAI approaches directed at image classifiers, I show that most of them explain with samples. If my analysis is correct, it will be beneficial if such explainability methods use explanations that qualify as exemplars. (shrink)

Previous research in Explainable Artificial Intelligence (XAI) suggests that a main aim of explainability approaches is to satisfy specific interests, goals, expectations, needs, and demands regarding artificial systems (we call these “stakeholders' desiderata”) in a variety of contexts. However, the literature on XAI is vast, spreads out across multiple largely disconnected disciplines, and it often remains unclear how explainability approaches are supposed to achieve the goal of satisfying stakeholders' desiderata. This paper discusses the main classes of stakeholders calling for explainability (...) of artificial systems and reviews their desiderata. We provide a model that explicitly spells out the main concepts and relations necessary to consider and investigate when evaluating, adjusting, choosing, and developing explainability approaches that aim to satisfy stakeholders' desiderata. This model can serve researchers from the variety of different disciplines involved in XAI as a common ground. It emphasizes where there is interdisciplinary potential in the evaluation and the development of explainability approaches. (shrink)

In this paper I argue that the search for explainable models and interpretable decisions in AI must be reformulated in terms of the broader project of offering a pragmatic and naturalistic account of understanding in AI. Intuitively, the purpose of providing an explanation of a model or a decision is to make it understandable to its stakeholders. But without a previous grasp of what it means to say that an agent understands a model or a decision, the explanatory strategies will (...) lack a well-defined goal. Aside from providing a clearer objective for XAI, focusing on understanding also allows us to relax the factivity condition on explanation, which is impossible to fulfill in many machine learning models, and to focus instead on the pragmatic conditions that determine the best fit between a model and the methods and devices deployed to understand it. After an examination of the different types of understanding discussed in the philosophical and psychological literature, I conclude that interpretative or approximation models not only provide the best way to achieve the objectual understanding of a machine learning model, but are also a necessary condition to achieve post hoc interpretability. This conclusion is partly based on the shortcomings of the purely functionalist approach to post hoc interpretability that seems to be predominant in most recent literature. (shrink)

Explainable artificial intelligence (XAI) aims to help people understand black box algorithms, particularly of their outputs. But what are these explanations and when is one explanation better than another? The manipulationist definition of explanation from the philosophy of science offers good answers to these questions, holding that an explanation consists of a generalization that shows what happens in counterfactual cases. Furthermore, when it comes to explanatory depth this account holds that a generalization that has more abstract variables, is broader in (...) scope and/or more accurate is better. By applying these definitions and contrasting them with alternative definitions in the XAI literature I hope to help clarify what a good explanation is for AI. (shrink)

Explainable AI (xAI) methods are important for establishing trust in using black-box models. However, recent criticism has mounted against current xAI methods that they disagree, are necessarily false, and can be manipulated, which has started to undermine the deployment of black-box models. Rudin (2019) goes so far as to say that we should stop using black-box models altogether in high-stakes cases because xAI explanations ‘must be wrong’. However, strict fidelity to the truth is historically not a desideratum in science. Idealizations (...) – the intentional distortions introduced to scientific theories and models – are commonplace in the natural sciences and are seen as a successful scientific tool. Thus, it is not falsehood qua falsehood that is the issue. In this paper, I outline the need for xAI research to engage in idealization evaluation. Drawing on the use of idealizations in the natural sciences and philosophy of science, I introduce a novel framework for evaluating whether xAI methods engage in successful idealizations or deceptive explanations (SIDEs). SIDEs evaluates whether the limitations of xAI methods, and the distortions that they introduce, can be part of a successful idealization or are indeed deceptive distortions as critics suggest. I discuss the role that existing research can play in idealization evaluation and where innovation is necessary. Through a qualitative analysis we find that leading feature importance methods and counterfactual explanations are subject to idealization failure and suggest remedies for ameliorating idealization failure. (shrink)

Explainable artificial intelligence (XAI) seeks to produce explanations for those machine learning methods which are deemed opaque. However, there is considerable disagreement about what this means and how to achieve it. Authors disagree on what should be explained (topic), to whom something should be explained (stakeholder), how something should be explained (instrument), and why something should be explained (goal). In this paper, I employ insights from means-end epistemology to structure the field. According to means-end epistemology, different means ought to be (...) rationally adopted to achieve different epistemic ends. Applied to XAI, different topics, stakeholders, and goals thus require different instruments. I call this the means-end account of XAI. The means-end account has a descriptive and a normative component: on the one hand, I show how the specific means-end relations give rise to a taxonomy of existing contributions to the field of XAI; on the other hand, I argue that the suitability of XAI methods can be assessed by analyzing whether they are prescribed by a given topic, stakeholder, and goal. (shrink)

Necessity and sufficiency are the building blocks of all successful explanations. Yet despite their importance, these notions have been conceptually underdeveloped and inconsistently applied in explainable artificial intelligence (XAI), a fast-growing research area that is so far lacking in firm theoretical foundations. Building on work in logic, probability, and causality, we establish the central role of necessity and sufficiency in XAI, unifying seemingly disparate methods in a single formal framework. We provide a sound and complete algorithm for computing explanatory factors (...) with respect to a given context, and demonstrate its flexibility and competitive performance against state of the art alternatives on various tasks. (shrink)

This paper critically investigates the explainable artificial intelligence (XAI) project. I analyze the word “explain” in XAI and the theory of explanation and identify the discrepancy between the meaning of the explanation claimed to be necessary and that which is actually presented. After summarizing the history of AI related to explainability, I argue that American philosophy in the 1900s operated in the background of said history. I then extract the meaning of explanation in view of XAI, to elucidate the relationship (...) among AI, logic, and the theory of explanation. In so doing, I aim to reveal DARPA’s surreptitious definitional retreat in terms of its contents and formal fallacy of sophisma figurae dictionis, drawing from Kant’s paralogism. I conclude that this intentional fallacy preexists the XAI project and that presumptuous use of reason, which Kant criticizes, is underlying. (shrink)

Counterfactual reasoning -- envisioning hypothetical scenarios, or possible worlds, where some circumstances are different from what (f)actually occurred (counter-to-fact) -- is ubiquitous in human cognition. Conventionally, counterfactually-altered circumstances have been treated as "small miracles" that locally violate the laws of nature while sharing the same initial conditions. In Pearl's structural causal model (SCM) framework this is made mathematically rigorous via interventions that modify the causal laws while the values of exogenous variables are shared. In recent years, however, this purely interventionist (...) account of counterfactuals has increasingly come under scrutiny from both philosophers and psychologists. Instead, they suggest a backtracking account of counterfactuals, according to which the causal laws remain unchanged in the counterfactual world; differences to the factual world are instead "backtracked" to altered initial conditions (exogenous variables). In the present work, we explore and formalise this alternative mode of counterfactual reasoning within the SCM framework. Despite ample evidence that humans backtrack, the present work constitutes, to the best of our knowledge, the first general account and algorithmisation of backtracking counterfactuals. We discuss our backtracking semantics in the context of related literature and draw connections to recent developments in explainable artificial intelligence (XAI). (shrink)

Explainable artificial intelligence (XAI) is an emerging, multidisciplinary field of research that seeks to develop methods and tools for making AI systems more explainable or interpretable. XAI researchers increasingly recognise explainability as a context-, audience- and purpose-sensitive phenomenon, rather than a single well-defined property that can be directly measured and optimised. However, since there is currently no overarching definition of explainability, this poses a risk of miscommunication between the many different researchers within this multidisciplinary space. This is the problem we (...) seek to address in this paper. We outline a framework, called Explanatory Pragmatism, which we argue has two attractive features. First, it allows us to conceptualise explainability in explicitly context-, audience- and purpose-relative terms, while retaining a unified underlying definition of explainability. Second, it makes visible any normative disagreements that may underpin conflicting claims about explainability regarding the purposes for which explanations are sought. Third, it allows us to distinguish several dimensions of AI explainability. We illustrate this framework by applying it to a case study involving a machine learning model for predicting whether patients suffering disorders of consciousness were likely to recover consciousness. (shrink)

In a recent article, Erasmus, Brunet, and Fisher (2021) argue that Artificial Neural Networks (ANNs) are explainable. They survey four influential accounts of explanation: the Deductive-Nomological model, the Inductive-Statistical model, the Causal-Mechanical model, and the New-Mechanist model. They argue that, on each of these accounts, the features that make something an explanation is invariant with regard to the complexity of the explanans and the explanandum. Therefore, they conclude, the complexity of ANNs (and other Machine Learning models) does not make them (...) less explainable. In this reply, it is argued that Erasmus et al. left out one influential account of explanation from their discussion: the Unificationist model. It is argued that, on the Unificationist model, the features that makes something an explanation is sensitive to complexity. Therefore, on the Unificationist model, ANNs (and other Machine Learning models) are not explainable. It is emphasized that Erasmus et al.’s general strategy is correct. The literature on explainable Artificial Intelligence can benefit by drawing from philosophical accounts of explanation. However, philosophical accounts of explanation do not settle the problem of whether ANNs are explainable because they do not unanimously declare that explanation is invariant with regard to complexity. (shrink)

Many AI systems that make important decisions are black boxes: how they function is opaque even to their developers. This is due to their high complexity and to the fact that they are trained rather than programmed. Efforts to alleviate the opacity of black box systems are typically discussed in terms of transparency, interpretability, and explainability. However, there is little agreement about what these key concepts mean, which makes it difficult to adjudicate the success or promise of opacity alleviation methods. (...) I argue for a unified account of these key concepts that treats the concept of understanding as fundamental. This allows resources from the philosophy of science and the epistemology of understanding to help guide opacity alleviation efforts. A first significant benefit of this understanding account is that it defuses one of the primary, in-principle objections to post hoc explainable AI (XAI) methods. This “rationalization objection” argues that XAI methods provide mere rationalizations rather than genuine explanations. This is because XAI methods involve using a separate “explanation” system to approximate the original black box system. These explanation systems function in a completely different way than the original system, yet XAI methods make inferences about the original system based on the behavior of the explanation system. I argue that, if we conceive of XAI methods as idealized scientific models, this rationalization worry is dissolved. Idealized scientific models misrepresent their target phenomena, yet are capable of providing significant and genuine understanding of their targets. (shrink)

Artificial intelligence (AI) systems are increasingly adopted to make decisions in domains such as business, education, health care, and criminal justice. However, such algorithmic decision systems can have prevalent biases against marginalized social groups and undermine social justice. Explainable artificial intelligence (XAI) is a recent development aiming to make an AI system’s decision processes less opaque and to expose its problematic biases. This paper argues against technical XAI, according to which the detection and interpretation of algorithmic bias can be handled (...) more or less independently by technical experts who specialize in XAI methods. Drawing on resources from feminist epistemology, we show why technical XAI is mistaken. Specifically, we demonstrate that the proper detection of algorithmic bias requires relevant interpretive resources, which can only be made available, in practice, by actively involving a diverse group of stakeholders. Finally, we suggest how feminist theories can help shape integrated XAI: an inclusive social-epistemic process that facilitates the amelioration of algorithmic bias. (shrink)

The operations of deep networks are widely acknowledged to be inscrutable. The growing field of Explainable AI has emerged in direct response to this problem. However, owing to the nature of the opacity in question, XAI has been forced to prioritise interpretability at the expense of completeness, and even realism, so that its explanations are frequently interpretable without being underpinned by more comprehensive explanations faithful to the way a network computes its predictions. While this has been taken to be a (...) shortcoming of the field of XAI, I argue that it is broadly the right approach to the problem. (shrink)

In this paper, we contribute to research on enterprise artificial intelligence (AI), specifically to organizations improving the customer experiences and their internal processes through using the type of AI called machine learning (ML). Many organizations are struggling to get enough value from their AI efforts, and part of this is related to the area of explainability. The need for explainability is especially high in what is called black-box ML models, where decisions are made without anyone understanding how an AI reached (...) a particular decision. This opaqueness creates a user need for explanations. Therefore, researchers and designers create different versions of so-called eXplainable AI (XAI). However, the demands for XAI can reduce the accuracy of the predictions the AI makes, which can reduce the perceived usefulness of the AI solution, which, in turn, reduces the interest in designing the organizational task structure to benefit from the AI solution. Therefore, it is important to ensure that the need for XAI is as low as possible. In this paper, we demonstrate how to achieve this by optimizing the task structure according to sociotechnical systems design principles. Our theoretical contribution is to the underexplored field of the intersection of AI design and organizational design. We find that explainability goals can be divided into two groups, pattern goals and experience goals, and that this division is helpful when defining the design process and the task structure that the AI solution will be used in. Our practical contribution is for AI designers who include organizational designers in their teams, and for organizational designers who answer that challenge. (shrink)

In the past few years, machine learning (ML) tools have been implemented with success in the medical context. However, several practitioners have raised concerns about the lack of transparency—at the algorithmic level—of many of these tools; and solutions from the field of explainable AI (XAI) have been seen as a way to open the ‘black box’ and make the tools more trustworthy. Recently, Alex London has argued that in the medical context we do not need machine learning tools to be (...) interpretable at the algorithmic level to make them trustworthy, as long as they meet some strict empirical desiderata. In this paper, we analyse and develop London’s position. In particular, we make two claims. First, we claim that London’s solution to the problem of trust can potentially address another problem, which is how to evaluate the reliability of ML tools in medicine for regulatory purposes. Second, we claim that to deal with this problem, we need to develop London’s views by shifting the focus from the opacity of algorithmic details to the opacity of the way in which ML tools are trained and built. We claim that to regulate AI tools and evaluate their reliability, agencies need an explanation of how ML tools have been built, which requires documenting and justifying the technical choices that practitioners have made in designing such tools. This is because different algorithmic designs may lead to different outcomes, and to the realization of different purposes. However, given that technical choices underlying algorithmic design are shaped by value-laden considerations, opening the black box of the design process means also making transparent and motivating (technical and ethical) values and preferences behind such choices. Using tools from philosophy of technology and philosophy of science, we elaborate a framework showing how an explanation of the training processes of ML tools in medicine should look like. (shrink)

Autonomous artificial intelligence (AI) systems can lead to unpredictable behavior causing loss or damage to individuals. Intricate questions must be resolved to establish how courts determine liability. Until recently, understanding the inner workings of “black boxes” has been exceedingly difficult; however, the use of Explainable Artificial Intelligence (XAI) would help simplify the complex problems that can occur with autonomous AI systems. In this context, this article seeks to provide technical explanations that can be given by XAI, and to show how (...) suitable explanations for liability can be reached in court. It provides an analysis of whether existing liability frameworks, in both civil and common law tort systems, with the support of XAI, can address legal concerns related to AI. Lastly, it claims their further development and adoption should allow AI liability cases to be decided under current legal and regulatory rules until new liability regimes for AI are enacted. (shrink)

Many ethical frameworks require artificial intelligence (AI) systems to be explainable. Explainable AI (XAI) models are frequently tested for their adequacy in user studies. Since different people may have different explanatory needs, it is important that participant samples in user studies are large enough to represent the target population to enable generalizations. However, it is unclear to what extent XAI researchers reflect on and justify their sample sizes or avoid broad generalizations across people. We analyzed XAI user studies (N = (...) 220) published between 2012 and 2022. Most studies did not offer rationales for their sample sizes. Moreover, most papers generalized their conclusions beyond their target population, and there was no evidence that broader conclusions in quantitative studies were correlated with larger samples. These methodological problems can impede evaluations of whether XAI systems implement the explainability called for in ethical frameworks. We outline principles for more inclusive XAI user studies. (shrink)

In this paper I argue that the search for explainable models and interpretable decisions in AI must be reformulated in terms of the broader project of offering a pragmatic and naturalistic account of understanding in AI. Intuitively, the purpose of providing an explanation of a model or a decision is to make it understandable to its stakeholders. But without a previous grasp of what it means to say that an agent understands a model or a decision, the explanatory strategies will (...) lack a well-defined goal. Aside from providing a clearer objective for XAI, focusing on understanding also allows us to relax the factivity condition on explanation, which is impossible to fulfill in many machine learning models, and to focus instead on the pragmatic conditions that determine the best fit between a model and the methods and devices deployed to understand it. After an examination of the different types of understanding discussed in the philosophical and psychological literature, I conclude that interpretative or approximation models not only provide the best way to achieve the objectual understanding of a machine learning model, but are also a necessary condition to achieve post hoc interpretability. This conclusion is partly based on the shortcomings of the purely functionalist approach to post hoc interpretability that seems to be predominant in most recent literature. (shrink)

Modern explainable AI methods remain far from providing human-like answers to ‘why’ questions, let alone those that satisfactorily agree with human-level understanding. Instead, the results that such methods provide boil down to sets of causal attributions. Currently, the choice of accepted attributions rests largely, if not solely, on the explainee’s understanding of the quality of explanations. The paper argues that such decisions may be transferred from a human to an XAI agent, provided that its machine-learning algorithms perform genuinely abductive inferences. (...) The paper outlines the key predicament in the current inductive paradigm of ML and the associated XAI techniques, and sketches the desiderata for a truly participatory, second-generation XAI, which is endowed with abduction. (shrink)

A widespread need to explain the behavior and outcomes of AI-based systems has emerged, due to their ubiquitous presence. Thus, providing renewed momentum to the relatively new research area of eXplainable AI (XAI). Nowadays, the importance of XAI lies in the fact that the increasing control transference to this kind of system for decision making -or, at least, its use for assisting executive stakeholders- already affects many sensitive realms (as in Politics, Social Sciences, or Law). The decision-making power handover to (...) opaque AI systems makes mandatory explaining those, primarily in application scenarios where the stakeholders are unaware of both the high technology applied and the basic principles governing the technological solutions. The issue should not be reduced to a merely technical problem; the explainer would be compelled to transmit richer knowledge about the system (including its role within the informational ecosystem where he/she works). To achieve such an aim, the explainer could exploit, if necessary, practices from other scientific and humanistic areas. The first aim of the paper is to emphasize and justify the need for a multidisciplinary approach that is beneficiated from part of the scientific and philosophical corpus on Explaining, underscoring the particular nuances of the issue within the field of Data Science. The second objective is to develop some arguments justifying the authors’ bet by a more relevant role of ideas inspired by, on the one hand, formal techniques from Knowledge Representation and Reasoning, and on the other hand, the modeling of human reasoning when facing the explanation. This way, explaining modeling practices would seek a sound balance between the pure technical justification and the explainer-explainee agreement. (shrink)

The article addresses the problem of identifying methods to develop the ability of artificial intelligence (AI) systems to provide explanations for their findings. This issue is not new, but, nowadays, the increasing complexity of AI systems is forcing scientists to intensify research in this direction. Modern neural networks contain hundreds of layers of neurons. The number of parameters of these networks reaches trillions, genetic algorithms generate thousands of generations of solutions, and the semantics of AI models become more complicated, going (...) to the quantum and non-local levels. The world’s leading companies are investing heavily in creating explainable AI (XAI). However, the result is still unsatisfactory: a person often cannot understand the “explanations” of AI because the latter makes decisions differently than a person, and perhaps because a good explanation is impossible within the framework of the classical AI paradigm. AI faced a similar problem 40 years ago when expert systems contained only a few hundred logical production rules. The problem was then solved by complicating the logic and building added knowledge bases to explain the conclusions given by AI. At present, other approaches are needed, primarily those that consider the external environment and the subjectivity of AI systems. This work focuses on solving this problem by immersing AI models in the social and economic environment, building ontologies of this environment, taking into account a user profile and creating conditions for purposeful convergence of AI solutions and conclusions to user-friendly goals. (shrink)

We argue that explainable artificial intelligence (XAI), specifically reason-giving XAI, often constitutes the most suitable way of ensuring that someone can properly be held responsible for decisions that are based on the outputs of artificial intelligent (AI) systems. We first show that, to close moral responsibility gaps (Matthias 2004), often a human in the loop is needed who is directly responsible for particular AI-supported decisions. Second, we appeal to the epistemic condition on moral responsibility to argue that, in order to (...) be responsible for her decision, the human in the loop has to have an explanation available of the system’s recommendation. Reason explanations are especially well-suited to this end and we examine whether – and how – it might be possible to make such explanations fit with AI systems. We support our claims by focusing on a case of disagreement between human in the loop and AI system. (shrink)

There has been a push in recent years to provide better explanations for how AIs make their decisions. Most of this push has come from the ethical concerns that go hand in hand with AIs making decisions that affect humans. Outside of the strictly ethical concerns that have prompted the study of explainable AIs (XAIs), there has been research interest in the mere possibility of creating XAIs in various domains. In general, the more accurate we make our models the harder (...) they are to explain. Go playing AIs like AlphaGo and KataGo provide fantastic examples of this phenomenon. In this paper, I discuss a non-exhaustive list of the leading theories of explanation and what each of these theories would say about the explainability of AI-played moves of Go. Finally, I consider the possibility of ever explaining AI-played Go moves in a way that meets the four principles of XAI. I conclude, somewhat pessimistically, that Go is not as imminently explainable as other domains. As such, the probability of having an XAI for Go that meets the four principles is low. (shrink)

There is general consensus that explainable artificial intelligence is valuable, but there is significant divergence when we try to articulate why, exactly, it is desirable. This question must be distinguished from two other kinds of questions asked in the XAI literature that are sometimes asked and addressed simultaneously. The first and most obvious is the ‘how’ question—some version of: ‘how do we develop technical strategies to achieve XAI?’ Another question is specifying what kind of explanation is worth having in the (...) first place. As difficult and important as the challenges are in answering these questions, they are distinct from a third question: why do we want XAI at all? There is vast literature on this question as well, but I wish to explore a different kind of answer. The most obvious way to answer this question is by describing a desirable outcome that would likely be achieved with the right kind of explanation, which would make the explanation valuable instrumentally. That is, XAI is desirable to attain some other value, such as fairness, trust, accountability, or governance. This family of arguments is obviously important, but I argue that explanations are also intrinsically valuable, because unexplainable systems can be dehumanizing. I argue that there are at least three independently valid versions of this kind of argument: an argument from participation, from knowledge, and from actualization. Each of these arguments that XAI is intrinsically valuable is independently compelling, in addition to the more obvious instrumental benefits of XAI. (shrink)

Implicit stochastic models, including both ‘deep neural networks’ (dNNs) and the more recent unsupervised foundational models, cannot be explained. That is, it cannot be determined how they work, because the interactions of the millions or billions of terms that are contained in their equations cannot be captured in the form of a causal model. Because users of stochastic AI systems would like to understand how they operate in order to be able to use them safely and reliably, there has emerged (...) a new field called ‘explainable AI’ (XAI). When we examine the XAI literature, however, it becomes apparent that its protagonists have redefined the term ‘explanation’ to mean something else, namely: ‘interpretation’. Interpretations are indeed sometimes possible, but we show that they give at best only a subjective understanding of how a model works. We propose an alternative to XAI, namely certified AI (CAI), and describe how an AI can be specified, realized, and tested in order to become certified. The resulting approach combines ontologies and formal logic with statistical learning to obtain reliable AI systems which can be safely used in technical applications. (shrink)

We argue that artificial networks are explainable and offer a novel theory of interpretability. Two sets of conceptual questions are prominent in theoretical engagements with artificial neural networks, especially in the context of medical artificial intelligence: Are networks explainable, and if so, what does it mean to explain the output of a network? And what does it mean for a network to be interpretable? We argue that accounts of “explanation” tailored specifically to neural networks have ineffectively reinvented the wheel. In (...) response to, we show how four familiar accounts of explanation apply to neural networks as they would to any scientific phenomenon. We diagnose the confusion about explaining neural networks within the machine learning literature as an equivocation on “explainability,” “understandability” and “interpretability.” To remedy this, we distinguish between these notions, and answer by offering a theory and typology of interpretation in machine learning. Interpretation is something one does to an explanation with the aim of producing another, more understandable, explanation. As with explanation, there are various concepts and methods involved in interpretation: Total or Partial, Global or Local, and Approximative or Isomorphic. Our account of “interpretability” is consistent with uses in the machine learning literature, in keeping with the philosophy of explanation and understanding, and pays special attention to medical artificial intelligence systems. (shrink)

The counterfactual approach to explainable AI (XAI) seeks to provide understanding of AI systems through the provision of counterfactual explanations. In a recent systematic review, Chou et al. (Inform Fus 81:59–83, 2022) argue that the counterfactual approach does not clearly provide causal understanding. They diagnose the problem in terms of the underlying framework within which the counterfactual approach has been developed. To date, the counterfactual approach has not been developed in concert with the approach for specifying causes developed by Pearl (...) (Causality: Models, reasoning, and inference. Cambridge University Press, 2000) and Woodward (Making things happen: A theory of causal explanation. Oxford University Press, 2003). In this paper, I build on Chou et al.’s work by applying the Pearl-Woodward approach. I argue that the standard counterfactual approach to XAI is capable of delivering causal understanding, but that there are limitations on its capacity to do so. I suggest a way to overcome these limitations. (shrink)

For synergistic interactions between humans and artificial intelligence (AI) systems, AI outputs often need to be explainable to people. Explainable AI (XAI) systems are commonly tested in human user studies. However, whether XAI researchers consider potential cultural differences in human explanatory needs remains unexplored. We highlight psychological research that found significant differences in human explanations between many people from Western, commonly individualist countries and people from non-Western, often collectivist countries. We argue that XAI research currently overlooks these variations and that (...) many popular XAI designs implicitly and problematically assume that Western explanatory needs are shared cross-culturally. Additionally, we systematically reviewed over 200 XAI user studies and found that most studies did not consider relevant cultural variations, sampled only Western populations, but drew conclusions about human-XAI interactions more generally. We also analyzed over 30 literature reviews of XAI studies. Most reviews did not mention cultural differences in explanatory needs or flag overly broad cross-cultural extrapolations of XAI user study results. Combined, our analyses provide evidence of a cultural bias toward Western populations in XAI research, highlighting an important knowledge gap regarding how culturally diverse users may respond to widely used XAI systems that future work can and should address. (shrink)

A recent article by Herzog provides a much-needed integration of ethical and epistemological arguments in favor of explicable AI in medicine. In this short piece, I suggest a way in which its epistemological intuition of XAI as “explanatory interface” can be further developed to delineate the relation between AI tools and scientific research.

Recent developments in artificial intelligence (AI) promise cognitive optimization in many areas of our lives, ranging from automated decision-making to superintelligence. In a predominant narrative, the black-box of machine learning systems is identified as one of the biggest obstacles from an epistemic point of view. The problem is expected to be solved by algorithmic counteractions emerging from the field of explainable artificial intelligence (XAI). However, deeper questions about a meaningful cognitive division of labor between AI algorithms and human actors who (...) use these systems are vastly ignored. The article aims to replace the dominant anthropomorphizing metaphors by a reconceptualization based on philosophy of technology. The new framework allows to interpret AI technology as an epistemic tool, which co-shapes a specific milieu of reflection and which is based on an epistemic model of automated knowledge production. This re-conceptualization offers a new perspective on AI technologies, where epistemic empowerment and epistemic sovereignty are in the center. (shrink)

This paper develops an account of the opacity problem in medical machine learning (ML). Guided by pragmatist assumptions, I argue that opacity in ML models is problematic insofar as it potentially undermines the achievement of two key purposes: ensuring generalizability and optimizing clinician–machine decision-making. Three opacity amelioration strategies are examined, with explainable artificial intelligence (XAI) as the predominant approach, challenged by two revisionary strategies in the form of reliabilism and the interpretability by design. Comparing the three strategies, I argue that (...) interpretability by design is most promising to overcome opacity in medical ML. Looking beyond the individual opacity amelioration strategies, the paper also contributes to a deeper understanding of the problem space and the solution space regarding opacity in medical ML. (shrink)

Artificial Intelligence (AI) pervades humanity in 2022, and it is notoriously difficult to understand how certain aspects of it work. There is a movement—_Explainable_ Artificial Intelligence (XAI)—to develop new methods for explaining the behaviours of AI systems. We aim to highlight one important philosophical significance of XAI—it has a role to play in the elimination of vagueness. To show this, consider that the use of AI in what has been labeled _surveillance capitalism_ has resulted in humans quickly gaining the capability (...) to identify and classify most of the occasions in which languages are used. We show that the knowability of this information is incompatible with what a certain theory of vagueness—_epistemicism_—says about vagueness. We argue that one way the epistemicist could respond to this threat is to claim that this process brought about the end of vagueness. However, we suggest an alternative interpretation, namely that epistemicism is false, but there is a weaker doctrine we dub _technological epistemicism_, which is the view that vagueness is due to ignorance of linguistic usage, but the ignorance can be overcome. The idea is that knowing more of the relevant data and how to process it enables us to know the semantic values of our words and sentences with higher confidence and precision. Finally, we argue that humans are probably not going to believe what future AI algorithms tell us about the sharp boundaries of our vague words unless the AI involved can be explained in terms understandable by humans. That is, if people are going to accept that AI can tell them about the sharp boundaries of the meanings of their words, then it is going to have to be XAI. (shrink)

Businesses increasingly rely on algorithms that are data-trained sets of decision rules (i.e., the output of the processes often called “machine learning”) and implement decisions with little or no human intermediation. In this article, we provide a philosophical foundation for the claim that algorithmic decision-making gives rise to a “right to explanation.” It is often said that, in the digital era, informed consent is dead. This negative view originates from a rigid understanding that presumes informed consent is a static and (...) complete transaction. Such a view is insufficient, especially when data are used in a secondary, noncontextual, and unpredictable manner—which is the inescapable nature of advanced artificial intelligence systems. We submit that an alternative view of informed consent—as an assurance of trust for incomplete transactions—allows for an understanding of why the rationale of informed consent already entails a right to ex post explanation. (shrink)

Advancements and deployments of AI-based systems, especially Deep Learning-driven generative language models, have accomplished impressive results over the past few years. Nevertheless, these remarkable achievements are intertwined with a related fear that such technologies might lead to a general relinquishing of our lives’s control to AIs. This concern, which also motivates the increasing interest in the eXplainable Artificial Intelligence (XAI) research field, is mostly caused by the opacity of the output of deep learning systems and the way that it is (...) generated, which is largely obscure to laypeople. A dialectical interaction with such systems may enhance the users’ understanding and build a more robust trust towards AI. Commonly employed as specific formalisms for modelling intra-agent communications, dialogue games prove to be useful tools to rely upon when dealing with user’s explanation needs. The literature already offers some dialectical protocols that expressly handle explanations and their delivery. This paper fully formalises the novel Explanation–Question–Response (EQR) dialogue and its properties, whose main purpose is to provide satisfactory information (i.e., justified according to argumentative semantics) whilst ensuring a simplified protocol, in comparison with other existing approaches, for humans and artificial agents. (shrink)

Necessity and sufficiency are the building blocks of all successful explanations. Yet despite their importance, these notions have been conceptually underdeveloped and inconsistently applied in explainable artificial intelligence, a fast-growing research area that is so far lacking in firm theoretical foundations. In this article, an expanded version of a paper originally presented at the 37th Conference on Uncertainty in Artificial Intelligence, we attempt to fill this gap. Building on work in logic, probability, and causality, we establish the central role of (...) necessity and sufficiency in XAI, unifying seemingly disparate methods in a single formal framework. We propose a novel formulation of these concepts, and demonstrate its advantages over leading alternatives. We present a sound and complete algorithm for computing explanatory factors with respect to a given context and set of agentive preferences, allowing users to identify necessary and sufficient conditions for desired outcomes at minimal cost. Experiments on real and simulated data confirm our method’s competitive performance against state of the art XAI tools on a diverse array of tasks. (shrink)

When artificial intelligence (AI) is used to make high-stakes decisions, some worry that this will create a morally troubling responsibility gap—that is, a situation in which nobody is morally responsible for the actions and outcomes that result. Since the responsibility gap might be thought to result from individuals lacking knowledge of the future behavior of AI systems, it can be and has been suggested that deploying explainable artificial intelligence (XAI) techniques will help us to avoid it. These techniques provide humans (...) with certain forms of understanding of the systems in question. In this paper, I consider whether existing XAI techniques can indeed close the responsibility gap. I identify a number of significant limits to their ability to do so. Ensuring that responsibility for AI-assisted outcomes is maintained may require using different techniques in different circumstances, and potentially also developing new techniques that can avoid each of the issues identified. (shrink)

Artificial intelligence (AI) has become nearly ubiquitous in modern society, from components of mobile applications to medical support systems, and everything in between. In societally impactful systems imbued with AI, there has been increasing concern related to opaque AI, that is, artificial intelligence where it is unclear how or why certain decisions are reached. This has led to a recent boom in research on “explainable AI” (XAI), or approaches to making AI more explainable and understandable to human users. In the (...) military domain, numerous bodies have argued that autonomous and AI-enabled weapon systems ought not incorporate unexplainable AI, with the International Committee of the Red Cross and the United States Department of Defense both explicitly including explainability as a relevant factor in the development and use of such systems. In this article, I present a cautiously critical assessment of this view, arguing that explainability will be irrelevant for many current and near-future autonomous systems in the military (which do not incorporate any AI), that it will be trivially incorporated into most military systems which do possess AI (as these generally possess simpler AI systems), and that for those systems with genuinely opaque AI, explainability will prove to be of more limited value than one might imagine. In particular, I argue that explainability, while indeed a virtue in design, is a virtue aimed primarily at designers and troubleshooters of AI-enabled systems, but is far less relevant for users and handlers actually deploying these systems. I further argue that human–machine teaming is a far more important element of responsibly using AI for military purposes, adding that explainability may undermine efforts to improve human–machine teamings by creating a prima facie sense that the AI, due to its explainability, may be utilized with little (or less) potential for mistakes. I conclude by clarifying that the arguments are not against XAI in the military, but are instead intended as a caution against over-inflating the value of XAI in this domain, or ignoring the limitations and potential pitfalls of this approach. (shrink)

Advancements in machine learning have fuelled the popularity of using AI decision algorithms in procedures such as bail hearings, medical diagnoses and recruitment. Academic articles, policy texts, and popularizing books alike warn that such algorithms tend to be opaque: they do not provide explanations for their outcomes. Building on a causal account of transparency and opacity as well as recent work on the value of causal explanation, I formulate a moral concern for opaque algorithms that is yet to receive a (...) systematic treatment in the literature: when such algorithms are used in life-changing decisions, they can obstruct us from effectively shaping our lives according to our goals and preferences, thus undermining our autonomy. I argue that this concern deserves closer attention as it furnishes the call for transparency in algorithmic decision-making with both new tools and new challenges. (shrink)

In a recent reply to our article, “What is Interpretability?,” Prasetya argues against our position that artificial neural networks are explainable. It is claimed that our indefeasibility thesis—that adding complexity to an explanation of a phenomenon does not make the phenomenon any less explainable—is false. More precisely, Prasetya argues that unificationist explanations are defeasible to increasing complexity, and thus, we may not be able to provide such explanations of highly complex AI models. The reply highlights an important lacuna in our (...) original paper, the omission of the unificationist account of explanation, and affords us the opportunity to respond. Here, we argue that artificial neural networks are explainable in a way that should satisfy unificationists and that interpretability methods present ways in which ML theories can achieve unification. (shrink)

ABSTRACT Global biodiversity declines are increasingly recognized as profound ecological and social crises. In areas subject to colonialization, these declines have advanced in lockstep with settler colonialism and imposition of centralized resource management by settler states. Many have suggested that resurgent Indigenous-led governance systems could help arrest these trends while advancing effective and socially just approaches to environmental interactions that benefit people and places alike. However, how dominant management and conservation approaches might be decolonized (i.e., how their underlying colonial structure (...) might be addressed, transformed, and replaced) is not always clear. Here, we describe a ‘Decolonial Model of Environmental Management and Conservation’ as an alternative paradigm to dominant approaches of conservation and management. The tenets of the model describe characteristics that might be expected of decolonized management, contrasted with those of dominant state-led approaches such as those embedded in the North American Model of Wildlife Conservation. The model does not prescribe how Indigenous governments or communities ought to govern their own territories, but instead offers insights into how external management and conservation agencies and practitioners might support (or stop impeding) Indigenous-led governance. We illustrate the model with a conservation ‘bright spot’: grizzly bear stewardship in the area now referred to as the Great Bear Rainforest in British Columbia, Canada, with a focus on work led by or in collaboration with, and within the territories of, the Haíɫzaqv, Kitasoo/xai’xais, Nuxalk, and Wuikinuxv First Nations. While acknowledging the important context-specific variability among place-based management and conservation applications, we also discuss the model’s broader applicability. (shrink)