This paper develops an account of how the implementation of ML models into healthcare settings requires revising the methodological apparatus of philosophical bioethics. On this account, ML models are cognitive interventions that provide decision-support to physicians and patients. Due to reliability issues, opaque reasoning processes, and information asymmetries, ML models pose inferential problems for them. These inferential problems lay the grounds for many ethical problems that currently claim centre-stage in the bioethical debate. Accordingly, this paper argues that the best way (...) to make progress in remedying these ethical problems is to distil their epistemic core and to identify appropriate epistemic norms as guardrails. The viability of this approach will be highlighted based on four key issues: trust, responsibility, paternalism, and fairness. In that respect, the paper also contributes to a more pronounced view of the specific methodological challenges for ethical theorizing, when the point of reference are complex statistical models that make predictions, rather than individual agents, acting for reasons. (shrink)

Technology regulation is one of the most important public policy issues facing society and governments at the present time, and further clarity could improve decision making in this complex and challenging area. Since the rise of the internet in the late 1990s, a number of approaches to technology regulation have been proposed, prompted by the associated changes in society, business and law that this development brought with it. However, over the past decade, the impact of technology has been profound and (...) the associated issues for government have extremely challenging, ranging across cyber security, artificial intelligence, and many other areas. To that end, this article introduces a Theory of Institutional Technology Actors and Norms (TITAN), a normatively informed and institutionally-based account of technology regulation. It focuses on the moral and legal (including regulatory) rights and responsibilities of the relevant actors and seeks to inform the development of regulation that is both fit for purpose, rights compliant and fair for all concerned. The account incorporates the perspectives of four key categories of groups in society: producers of technology, users of technology, government regulators, and normative policy shapers. (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)

When discussing AI alignment, we usually refer to the problem of teaching or training advanced autonomous AI systems to make decisions that are aligned with human values or preferences. Proponents of this approach believe it can be employed as means to stay in control over sophisticated intelligent systems, thus avoiding certain existential risks. We identify three general obstacles on the path to implementation of value alignment: a technological/technical obstacle, a normative obstacle, and a calibration problem. Presupposing, for the purposes of (...) this discussion, that the technical and normative problems are solved, we focus on the problem of how to calibrate a system, for a specific value, to be on a specific location within a spectrum stretching between righteous and normal or average human behavior. Calibration, or more specifically mis-calibration, also raises the issue of trustworthiness. If we cannot trust AI systems to perform tasks the way we intended, we would not use them on our roads and at our homes. In an era where we strive to construct autonomous machines endowed with common sense, reasoning abilities and a connection to the world, so they would be able to act in alignment with human values, such mis-calibrations can make the difference between trustworthy and untrustworthy systems. (shrink)

Without doctors being able to explain medical decisions to patients, I argue their use of black box AIs would erode the effective and respectful care they provide patients. In addition, I argue that physicians should use AI black boxes only for patients in dire straits, or when physicians use AI as a “co-pilot” (analogous to a spellchecker) but can independently confirm its accuracy. I respond to A.J. London’s objection that physicians already prescribe some drugs without knowing why they work.

Background Biomedicine, i.e. the application of basic sciences to medicine, has become the cornerstone for the study of etiopathogenesis and treatment of diseases. Biomedicine has enormously contributed to the progress of medicine and healthcare and has become the preferred approach to medical problems in the West. The developments in statistical inference and machine learning techniques have provided the foundation for personalised medicine where clinical management can be fully informed by biomedicine. The deployment of precision medicine may impact the autonomy and (...) self-normativity of the patients. Understanding the relationship between biomedicine and medical practice can help navigate the benefits and challenges offered by precision medicine. Methods Conventional content analysis was applied to “Le Normal and le Pathologique” (Canguilhem G. The Normal and the Pathological. Princeton: Princeton University Press; 1991) and further investigated with respect to its relationship with techne and precision medicine using PubMed and Google Scholar and the Standford Encyclopedia of Philosophy to search for the following keywords singularly or in combination: “Canguilhem”, “techne”, “episteme”, “precision medicine”, “machine learning AND medicine”. Results The Hippocratic concept of techne accounts for many characteristics of medical knowledge and practice. The advances of biomedicine, experimental medicine and, more recently, machine learning offer, in contrast, the model of a medicine based purely on episteme. I argue that Canguilhem medical epistemology establishes a framework where episteme and data-driven medicine is compatible with the promotion of patient’s autonomy and self-normativity. Conclusions Canguilhem’s medical epistemology orders the relationship of applied medicine with experimental sciences, ethics and social sciences. It provides guidance to define the scope of medicine and the boundaries of medicalization of healthy life. Finally, it sets an agenda for a safe implementation of machine learning in medicine. (shrink)

Artificial intelligence (AI) systems have demonstrated impressive performance across a variety of clinical tasks. However, notoriously, sometimes these systems are “black boxes.” The initial response in the literature was a demand for “explainable AI.” However, recently, several authors have suggested that making AI more explainable or “interpretable” is likely to be at the cost of the accuracy of these systems and that prioritizing interpretability in medical AI may constitute a “lethal prejudice.” In this paper, we defend the value of interpretability (...) in the context of the use of AI in medicine. Clinicians may prefer interpretable systems over more accurate black boxes, which in turn is sufficient to give designers of AI reason to prefer more interpretable systems in order to ensure that AI is adopted and its benefits realized. Moreover, clinicians may be justified in this preference. Achieving the downstream benefits from AI is critically dependent on how the outputs of these systems are interpreted by physicians and patients. A preference for the use of highly accurate black box AI systems, over less accurate but more interpretable systems, may itself constitute a form of lethal prejudice that may diminish the benefits of AI to—and perhaps even harm—patients. (shrink)

Artificial intelligence (AI) systems have demonstrated impressive performance across a variety of clinical tasks. However, notoriously, sometimes these systems are 'black boxes'. The initial response in the literature was a demand for 'explainable AI'. However, recently, several authors have suggested that making AI more explainable or 'interpretable' is likely to be at the cost of the accuracy of these systems and that prioritising interpretability in medical AI may constitute a 'lethal prejudice'. In this paper, we defend the value of interpretability (...) in the context of the use of AI in medicine. Clinicians may prefer interpretable systems over more accurate black boxes, which in turn is sufficient to give designers of AI reason to prefer more interpretable systems in order to ensure that AI is adopted and its benefits realised. Moreover, clinicians may be justified in this preference. Achieving the downstream benefits from AI is critically dependent on how the outputs of these systems are interpreted by physicians and patients. A preference for the use of highly accurate black box AI systems, over less accurate but more interpretable systems , may itself constitute a form of lethal prejudice that may diminish the benefits of AI to - and perhaps even harm - patients. (shrink)

The use of machine learning systems for decision-support in healthcare may exacerbate health inequalities. However, recent work suggests that algorithms trained on sufficiently diverse datasets could in principle combat health inequalities. One concern about these algorithms is that their performance for patients in traditionally disadvantaged groups exceeds their performance for patients in traditionally advantaged groups. This renders the algorithmic decisions unfair relative to the standard fairness metrics in machine learning. In this paper, we defend the permissible use of affirmative algorithms; (...) that is, algorithms trained on diverse datasets that perform better for traditionally disadvantaged groups. Whilst such algorithmic decisions may be unfair, the fairness of algorithmic decisions is not the appropriate locus of moral evaluation. What matters is the fairness of final decisions, such as diagnoses, resulting from collaboration between clinicians and algorithms. We argue that affirmative algorithms can permissibly be deployed provided the resultant final decisions are fair. (shrink)

Machine learning operates at the intersection of statistics and computer science. This raises the question as to its underlying methodology. While much emphasis has been put on the close link between the process of learning from data and induction, the falsificationist component of machine learning has received minor attention. In this paper, we argue that the idea of falsification is central to the methodology of machine learning. It is commonly thought that machine learning algorithms infer general prediction rules from past (...) observations. This is akin to a statistical procedure by which estimates are obtained from a sample of data. But machine learning algorithms can also be described as choosing one prediction rule from an entire class of functions. In particular, the algorithm that determines the weights of an artificial neural network operates by empirical risk minimization and rejects prediction rules that lack empirical adequacy. It also exhibits a behavior of implicit regularization that pushes hypothesis choice toward simpler prediction rules. We argue that taking both aspects together gives rise to a falsificationist account of artificial neural networks. (shrink)

Assistive systems based on Artificial Intelligence (AI) are bound to reshape decision-making in all areas of society. One of the most intricate challenges arising from their implementation in high-stakes environments such as medicine concerns their frequently unsatisfying levels of explainability, especially in the guise of the so-called black-box problem: highly successful models based on deep learning seem to be inherently opaque, resisting comprehensive explanations. This may explain why some scholars claim that research should focus on rendering AI systems understandable, rather (...) than explainable. Yet, there is a grave lack of agreement concerning these terms in much of the literature on AI. We argue that the seminal distinction made by the philosopher and physician Karl Jaspers between different types of explaining and understanding in psychopathology can be used to promote greater conceptual clarity in the context of Machine Learning (ML). Following Jaspers, we claim that explaining and understanding constitute multi-faceted epistemic approaches that should not be seen as mutually exclusive, but rather as complementary ones as in and of themselves they are necessarily limited. Drawing on the famous example of Watson for Oncology we highlight how Jaspers’ methodology translates to the case of medical AI. Classical considerations from the philosophy of psychiatry can therefore inform a debate at the centre of current AI ethics, which in turn may be crucial for a successful implementation of ethically and legally sound AI in medicine. (shrink)

There is a current debate about if, and in what sense, machine learning systems used in the medical context need to be explainable. Those arguing in favor contend these systems require post hoc explanations for each individual decision to increase trust and ensure accurate diagnoses. Those arguing against suggest the high accuracy and reliability of the systems is sufficient for providing epistemic justified beliefs without the need for explaining each individual decision. But, as we show, both solutions have limitations—and it (...) is unclear either address the epistemic worries of the medical professionals using these systems. We argue these systems do require an explanation, but an institutional explanation. These types of explanations provide the reasons why the medical professional should rely on the system in practice—that is, they focus on trying to address the epistemic concerns of those using the system in specific contexts and specific occasions. But ensuring that these institutional explanations are fit for purpose means ensuring the institutions designing and deploying these systems are transparent about the assumptions baked into the system. This requires coordination with experts and end-users concerning how it will function in the field, the metrics used to evaluate its accuracy, and the procedures for auditing the system to prevent biases and failures from going unaddressed. We contend this broader explanation is necessary for either post hoc explanations or accuracy scores to be epistemically meaningful to the medical professional, making it possible for them to rely on these systems as effective and useful tools in their practices. (shrink)

Philosophy Compass, Volume 17, Issue 6, June 2022.

Machine intelligence already helps medical staff with a number of tasks. Ethical decision-making, however, has not been handed over to computers. In this proof-of-concept study, we show how an algorithm based on Beauchamp and Childress’ prima-facie principles could be employed to advise on a range of moral dilemma situations that occur in medical institutions. We explain why we chose fuzzy cognitive maps to set up the advisory system and how we utilized machine learning to train it. We report on the (...) difficult task of operationalizing the principles of beneficence, non-maleficence and patient autonomy, and describe how we selected suitable input parameters that we extracted from a training dataset of clinical cases. The first performance results are promising, but an algorithmic approach to ethics also comes with several weaknesses and limitations. Should one really entrust the sensitive domain of clinical ethics to machine intelligence? (shrink)

In this paper, we use some elements of the philosophical theories of Wilfrid Sellars and Robert Brandom for examining the interactions between humans and machines. In particular, we adopt the concept of the space of reasons for analyzing the status of artificial intelligent agents. One could argue that AIAs, like the widely used recommendation systems, have already entered the space of reasons, since they seem to make knowledge claims that we use as premises for further claims. This, in turn, can (...) lead to a sense of alienation because AIAs do not quite play by the rules of the space of the reason. We, therefore, ask somewhat pointedly whether aliens have entered the space of reasons. A closer look reveals that it is a misconception to consider AIAs as being in the space of reasons. In fact, they should be seen as very sophisticated tools. Since these tools affect our own acting in the space of reasons, special regulations are required for their proper use. (shrink)

BackgroundMachine learning-based clinical decision support systems (ML_CDSS) are increasingly employed in various sectors of health care aiming at supporting clinicians’ practice by matching the characteristics of individual patients with a computerised clinical knowledge base. Some studies even indicate that ML_CDSS may surpass physicians’ competencies regarding specific isolated tasks. From an ethical perspective, however, the usage of ML_CDSS in medical practice touches on a range of fundamental normative issues. This article aims to add to the ethical discussion by using professionalisation theory (...) as an analytical lens for investigating how medical action at the micro level and the physician–patient relationship might be affected by the employment of ML_CDSS.Main textProfessionalisation theory, as a distinct sociological framework, provides an elaborated account of what constitutes client-related professional action, such as medical action, at its core and why it is more than pure expertise-based action. Professionalisation theory is introduced by presenting five general structural features of professionalised medical practice: (i) the patient has a concern; (ii) the physician deals with the patient’s concern; (iii) s/he gives assistance without patronising; (iv) s/he regards the patient in a holistic manner without building up a private relationship; and (v) s/he applies her/his general expertise to the particularities of the individual case. Each of these five key aspects are then analysed regarding the usage of ML_CDSS, thereby integrating the perspectives of professionalisation theory and medical ethics.ConclusionsUsing ML_CDSS in medical practice requires the physician to pay special attention to those facts of the individual case that cannot be comprehensively considered by ML_CDSS, for example, the patient’s personality, life situation or cultural background. Moreover, the more routinized the use of ML_CDSS becomes in clinical practice, the more that physicians need to focus on the patient’s concern and strengthen patient autonomy, for instance, by adequately integrating digital decision support in shared decision-making. (shrink)

In recent years, there has been a surge of high-profile publications on applications of artificial intelligence (AI) systems for medical diagnosis and prognosis. While AI provides various opportunities for medical practice, there is an emerging consensus that the existing studies show considerable deficits and are unable to establish the clinical benefit of AI systems. Hence, the view that the clinical benefit of AI systems needs to be studied in clinical trials—particularly randomised controlled trials (RCTs)—is gaining ground. However, an issue that (...) has been overlooked so far in the debate is that, compared with drug RCTs, AI RCTs require methodological adjustments, which entail ethical challenges. This paper sets out to develop a systematic account of the ethics of AI RCTs by focusing on the moral principles of clinical equipoise, informed consent and fairness. This way, the objective is to animate further debate on the (research) ethics of medical AI. (shrink)

The development of artificial intelligence (AI) in medicine raises fundamental ethical issues. As one example, AI systems in the field of mental health successfully detect signs of mental disorders, such as depression, by using data from social media. These AI depression detectors (AIDDs) identify users who are at risk of depression prior to any contact with the healthcare system. The article focuses on the ethical implications of AIDDs regarding affected users’ health-related autonomy. Firstly, it presents the (ethical) discussion of AI (...) in medicine and, specifically, in mental health. Secondly, two models of AIDDs using social media data and different usage scenarios are introduced. Thirdly, the concept of patient autonomy, according to Beauchamp and Childress, is critically discussed. Since this concept does not encompass the specific challenges linked with the digital context of AIDDs in social media sufficiently, the current analysis suggests, finally, an extended concept of health-related digital autonomy. (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)

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...

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)

There is widespread agreement that there should be a principle requiring that artificial intelligence be ‘explicable’. Microsoft, Google, the World Economic Forum, the draft AI ethics guidelines for the EU commission, etc. all include a principle for AI that falls under the umbrella of ‘explicability’. Roughly, the principle states that “for AI to promote and not constrain human autonomy, our ‘decision about who should decide’ must be informed by knowledge of how AI would act instead of us” :689–707, 2018). There (...) is a strong intuition that if an algorithm decides, for example, whether to give someone a loan, then that algorithm should be explicable. I argue here, however, that such a principle is misdirected. The property of requiring explicability should attach to a particular action or decision rather than the entity making that decision. It is the context and the potential harm resulting from decisions that drive the moral need for explicability—not the process by which decisions are reached. Related to this is the fact that AI is used for many low-risk purposes for which it would be unnecessary to require that it be explicable. A principle requiring explicability would prevent us from reaping the benefits of AI used in these situations. Finally, the explanations given by explicable AI are only fruitful if we already know which considerations are acceptable for the decision at hand. If we already have these considerations, then there is no need to use contemporary AI algorithms because standard automation would be available. In other words, a principle of explicability for AI makes the use of AI redundant. (shrink)

In an analysis of artificially intelligent systems for medical diagnostics and treatment planning we argue that patients should be able to exercise a right to withdraw from AI diagnostics and treatment planning for reasons related to (1) the physician’s role in the patients’ formation of and acting on personal preferences and values, (2) the bias and opacity problem of AI systems, and (3) rational concerns about the future societal effects of introducing AI systems in the health care sector.

Simple idealized models seem to provide more understanding than opaque, complex, and hyper-realistic models. However, an increasing number of scientists are going in the opposite direction by utilizing opaque machine learning models to make predictions and draw inferences, suggesting that scientists are opting for models that have less potential for understanding. Are scientists trading understanding for some other epistemic or pragmatic good when they choose a machine learning model? Or are the assumptions behind why minimal models provide understanding misguided? In (...) this paper, using the case of deep neural networks, I argue that it is not the complexity or black box nature of a model that limits how much understanding the model provides. Instead, it is a lack of scientific and empirical evidence supporting the link that connects a model to the target phenomenon that primarily prohibits understanding. (shrink)

I analyse an argument according to which medical artificial intelligence represents a threat to patient autonomy—recently put forward by Rosalind McDougall in the Journal of Medical Ethics. The argument takes the case of IBM Watson for Oncology to argue that such technologies risk disregarding the individual values and wishes of patients. I find three problems with this argument: it confuses AI with machine learning; it misses machine learning’s potential for personalised medicine through big data; it fails to distinguish between evidence-based (...) advice and decision-making within healthcare. I conclude that how much and which tasks we should delegate to machine learning and other technologies within healthcare and beyond is indeed a crucial question of our time, but in order to answer it, we must be careful in analysing and properly distinguish between the different systems and different delegated tasks. (shrink)

Artificial intelligence (AI) is increasingly being developed for use in medicine, including for diagnosis and in treatment decision making. The use of AI in medical treatment raises many ethical issues that are yet to be explored in depth by bioethicists. In this paper, I focus specifically on the relationship between the ethical ideal of shared decision making and AI systems that generate treatment recommendations, using the example of IBM’s Watson for Oncology. I argue that use of this type of system (...) creates both important risks and significant opportunities for promoting shared decision making. If value judgements are fixed and covert in AI systems, then we risk a shift back to more paternalistic medical care. However, if designed and used in an ethically informed way, AI could offer a potentially powerful way of supporting shared decision making. It could be used to incorporate explicit value reflection, promoting patient autonomy. In the context of medical treatment, we need value-flexible AI that can both respond to the values and treatment goals of individual patients and support clinicians to engage in shared decision making. (shrink)

The high predictive accuracy of contemporary machine learning-based AI systems has led some scholars to argue that, in certain cases, we should grant them epistemic expertise and authority over humans. This approach suggests that humans would have the epistemic obligation of relying on the predictions of a highly accurate AI system. Contrary to this view, in this work we claim that it is not possible to endow AI systems with a genuine account of epistemic expertise. In fact, relying on accounts (...) of expertise and authority from virtue epistemology, we show that epistemic expertise requires a relation with understanding that AI systems do not satisfy and intellectual abilities that these systems do not manifest. Further, following the Distribution Cognition theory and adapting an account by Croce on the virtues of collective epistemic agents to the case of human-AI interactions we show that, if an AI system is successfully appropriated by a human agent, a _hybrid_ epistemic agent emerges, which can become both an epistemic expert and an authority. Consequently, we claim that the aforementioned hybrid agent is the appropriate object of a discourse around trust in AI and the epistemic obligations that stem from its epistemic superiority. (shrink)

Through hypothetical scenarios, this paper analyses whether machine learning (ML) could resolve one of the main shortcomings present in Christopher Boorse’s Biostatistical Theory of health (BST). In doing so, it foregrounds the boundaries and challenges of employing ML in formulating a naturalist (i.e., prima facie value-free) definition of health. The paper argues that a sweeping dataist approach cannot fully make the BST truly naturalistic, as prior theories and values persist. It also points out that supervised learning introduces circularity, rendering it (...) incompatible with a naturalistic perspective. Additionally, it underscores the need for pre-existing auxiliary theories to assess results from unsupervised learning. It emphasizes the importance of understanding the epistemological entanglements between data and data processing methods to manage expectations about what data patterns can predict. In conclusion, the paper argues against delegating the final authority for defining complex concepts like health to AI systems, as it necessitates ethical judgment and capacities for deliberation that AI currently lacks. It also warns against granting creators and deployers of AI systems the discretionary authority to determine these definitions outside the wider social discussion, advocating for ongoing public engagement on normative notions. Failure to do so risks limiting individuals and collectives’ ability to shape a just digital future and diminishes their fundamental epistemic agency. (shrink)

Several proposals for moral enhancement would use AI to augment (auxiliary enhancement) or even supplant (exhaustive enhancement) human moral reasoning or judgment. Exhaustive enhancement proposals conceive AI as some self-contained oracle whose superiority to our own moral abilities is manifest in its ability to reliably deliver the ‘right’ answers to all our moral problems. We think this is a mistaken way to frame the project, as it presumes that we already know many things that we are still in the process (...) of working out, and reflecting on this fact reveals challenges even for auxiliary proposals that eschew the oracular approach. We argue there is nonetheless a substantial role that ‘AI mentors’ could play in our moral education and training. Expanding on the idea of an AI Socratic Interlocutor, we propose a modular system of multiple AI interlocutors with their own distinct points of view reflecting their training in a diversity of concrete wisdom traditions. This approach minimizes any risk of moral disengagement, while the existence of multiple modules from a diversity of traditions ensures pluralism is preserved. We conclude with reflections on how all this relates to the broader notion of moral transcendence implicated in the project of AI moral enhancement, contending it is precisely the whole concrete socio-technical system of moral engagement that we need to model if we are to pursue moral enhancement. (shrink)

We argue why interpretability should have primacy alongside empiricism for several reasons: first, if machine learning models are beginning to render some of the high-risk healthcare decisions instead of clinicians, these models pose a novel medicolegal and ethical frontier that is incompletely addressed by current methods of appraising medical interventions like pharmacological therapies; second, a number of judicial precedents underpinning medical liability and negligence are compromised when ‘autonomous’ ML recommendations are considered to be en par with human instruction in specific (...) contexts; third, explainable algorithms may be more amenable to the ascertainment and minimisation of biases, with repercussions for racial equity as well as scientific reproducibility and generalisability. We conclude with some reasons for the ineludible importance of interpretability, such as the establishment of trust, in overcoming perhaps the most difficult challenge ML will face in a high-stakes environment like healthcare: professional and public acceptance. (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)

Artificial intelligence (AI) systems are increasingly being used in healthcare, thanks to the high level of performance that these systems have proven to deliver. So far, clinical applications have focused on diagnosis and on prediction of outcomes. It is less clear in what way AI can or should support complex clinical decisions that crucially depend on patient preferences. In this paper, we focus on the ethical questions arising from the design, development and deployment of AI systems to support decision-making around (...) cardiopulmonary resuscitation and the determination of a patient’s Do Not Attempt to Resuscitate status (also known as code status). The COVID-19 pandemic has made us keenly aware of the difficulties physicians encounter when they have to act quickly in stressful situations without knowing what their patient would have wanted. We discuss the results of an interview study conducted with healthcare professionals in a university hospital aimed at understanding the status quo of resuscitation decision processes while exploring a potential role for AI systems in decision-making around code status. Our data suggest that (1) current practices are fraught with challenges such as insufficient knowledge regarding patient preferences, time pressure and personal bias guiding care considerations and (2) there is considerable openness among clinicians to consider the use of AI-based decision support. We suggest a model for how AI can contribute to improve decision-making around resuscitation and propose a set of ethically relevant preconditions—conceptual, methodological and procedural—that need to be considered in further development and implementation efforts. (shrink)

The focus of this paper are the ethical, legal and social challenges for ensuring the responsible use of “big brain data”—the recording, collection and analysis of individuals’ brain data on a large scale with clinical and consumer-directed neurotechnological devices. First, I highlight the benefits of big data and machine learning analytics in neuroscience for basic and translational research. Then, I describe some of the technological, social and psychological barriers for securing brain data from unwarranted access. In this context, I then (...) examine ways in which safeguards at the hardware and software level, as well as increasing “data literacy” in society, may enhance the security of neurotechnological devices and protect the privacy of personal brain data. Regarding ethical and legal ramifications of big brain data, I first discuss effects on the autonomy, the sense of agency and authenticity, as well as the self that may result from the interaction between users and intelligent, particularly closed-loop, neurotechnological devices. I then discuss the impact of the “datafication” in basic and clinical neuroscience research on the just distribution of resources and access to these transformative technologies. In the legal realm, I examine possible legal consequences that arises from the increasing abilities to decode brain states and their corresponding subjective phenomenological experiences on the hitherto inaccessible privacy of these information. Finally, I discuss the implications of big brain data for national and international regulatory policies and models of good data governance. (shrink)

Artificial intelligence (AI) is often described as crucial for making healthcare safer and more efficient. However, some studies point in the opposite direction, demonstrating how biases in AI cause inequalities and discrimination. As a result, a growing body of research suggests mitigation measures to avoid gender bias. Typically, mitigation measures address various stakeholders such as the industry, academia, and policy-makers. To the author’s knowledge, these have not undergone sociological analysis. The article fills this gap and explores five examples of mitigation (...) measures designed to counteract gender bias in AI within the healthcare sector. The rapid development of AI in healthcare plays a crucial role globally and must refrain from creating or reinforcing inequality and discrimination. In this effort, mitigation measures to avoid gender bias in AI in healthcare are central tools and, therefore, essential to explore from a social science perspective, including sociology. Sociologists have made valuable contributions to studying inequalities and disparities in AI. However, research has pointed out that more engagement is needed, specifically regarding bias in AI. While acknowledging the importance of these measures, the article suggests that they lack accountable agents for implementation and overlook potential implementation barriers such as resistance, power relations, and knowledge hierarchies. Recognizing the conditions where the mitigation measures are to be implemented is essential for understanding the potential challenges that may arise. Consequently, more studies are needed to explore the practical implementation of mitigation measures from a social science perspective and a systematic review of mitigation measures. (shrink)

Recently clinicians have become more reliant on technologies such as artificial intelligence (AI) and machine learning (ML) for effective and accurate diagnosis and prognosis of diseases, especially mental health disorders. These remarks, however, apply primarily to Europe, the USA, China and other technologically developed nations. Africa is yet to leverage the potential applications of AI and ML within the medical space. Sub-Saharan African countries are currently disadvantaged economically and infrastructure-wise. Yet precisely, these circumstances create significant opportunities for the deployment of (...) medical AI, which has already been deployed in some places in the continent. However, while AI and ML have come with enormous promises in Africa, there are still challenges when it comes to successfully applying AI and ML designed elsewhere within the African context, especially in diagnosing mental health disorders. We argue, in this paper, that there ought not to be a homogeneous/generic design of AI and ML used in diagnosing mental health disorders. Our claim is grounded on the premise that mental health disorders cannot be diagnosed solely on ‘factual evidence’ but on both factual evidence and value-laden judgements of what constitutes mental health disorders in sub-Saharan Africa. For ML to play a successful role in diagnosing mental health disorders in sub-Saharan African medical spaces, with a precise focus on South Africa, we allude that it ought to understand what sub-Saharan Africans consider as mental health disorders, that is, the value-laden judgements of some conditions. (shrink)

Making good decisions in extremely complex and difficult processes and situations has always been both a key task as well as a challenge in the clinic and has led to a large amount of clinical, legal and ethical routines, protocols and reflections in order to guarantee fair, participatory and up-to-date pathways for clinical decision-making. Nevertheless, the complexity of processes and physical phenomena, time as well as economic constraints and not least further endeavours as well as achievements in medicine and healthcare (...) continuously raise the need to evaluate and to improve clinical decision-making. This article scrutinises if and how clinical decision-making processes are challenged by the rise of so-called artificial intelligence-driven decision support systems (AI-DSS). In a first step, this article analyses how the rise of AI-DSS will affect and transform the modes of interaction between different agents in the clinic. In a second step, we point out how these changing modes of interaction also imply shifts in the conditions of trustworthiness, epistemic challenges regarding transparency, the underlying normative concepts of agency and its embedding into concrete contexts of deployment and, finally, the consequences for (possible) ascriptions of responsibility. Third, we draw first conclusions for further steps regarding a ‘meaningful human control’ of clinical AI-DSS. (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)

The concept of meaningful work has recently received increased attention in philosophy and other disciplines. However, the impact of the increasing robotization of the workplace on meaningful work has received very little attention so far. Doing work that is meaningful leads to higher job satisfaction and increased worker well-being, and some argue for a right to access to meaningful work. In this paper, we therefore address the impact of robotization on meaningful work. We do so by identifying five key aspects (...) of meaningful work: pursuing a purpose, social relationships, exercising skills and self-development, self-esteem and recognition, and autonomy. For each aspect, we analyze how the introduction of robots into the workplace may diminish or enhance the meaningfulness of work. We also identify a few ethical issues that emerge from our analysis. We conclude that robotization of the workplace can have both significant negative and positive effects on meaningful work. Our findings about ways in which robotization of the workplace can be a threat or opportunity for meaningful work can serve as the basis for ethical arguments for how to—and how not to—implement robots into workplaces. (shrink)

Within the ethical debate on Machine Learning-driven decision support systems (ML_CDSS), notions such as “human in the loop” or “meaningful human control” are often cited as being necessary for ethical legitimacy. In addition, ethical principles usually serve as the major point of reference in ethical guidance documents, stating that conflicts between principles need to be weighed and balanced against each other. Starting from a neo-Kantian viewpoint inspired by Onora O'Neill, this article makes a concrete suggestion of how to interpret the (...) role of the “human in the loop” and to overcome the perspective of rivaling ethical principles in the evaluation of AI in health care. We argue that patients should be perceived as “fellow workers” and epistemic partners in the interpretation of ML_CDSS outputs. We further highlight that a meaningful process of integrating (rather than weighing and balancing) ethical principles is most appropriate in the evaluation of medical AI. (shrink)

In light of recent advances in machine learning for medical applications, the automation of medical diagnostics is imminent. That said, before machine learning algorithms find their way into clinical practice, various problems at the epistemic level need to be overcome. In this paper, we discuss different sources of uncertainty arising for clinicians trying to evaluate the trustworthiness of algorithmic evidence when making diagnostic judgments. Thereby, we examine many of the limitations of current machine learning algorithms (with deep learning in particular) (...) and highlight their relevance for medical diagnostics. Among the problems we inspect are the theoretical foundations of deep learning (which are not yet adequately understood), the opacity of algorithmic decisions, and the vulnerabilities of machine learning models, as well as concerns regarding the quality of medical data used to train the models. Building on this, we discuss different desiderata for an uncertainty amelioration strategy that ensures that the integration of machine learning into clinical settings proves to be medically beneficial in a meaningful way. (shrink)

In their article ‘Who is afraid of black box algorithms? On the epistemological and ethical basis of trust in medical AI’, Durán and Jongsma discuss the epistemic and ethical challenges raised by black box algorithms in medical practice. The opacity of black box algorithms is an obstacle to the trustworthiness of their outcomes. Moreover, the use of opaque algorithms is not normatively justified in medical practice. The authors introduce a formalism, called computational reliabilism, which allows generating justified beliefs on the (...) algorithm reliability and trustworthy outcomes of artificial intelligence systems by means of epistemic warrants, called reliability indicators. However, they remark the need for reliability indicators specific to black box algorithms and that justified knowledge is not sufficient to justify normatively the actions of the physicians using medical AI systems. Therefore, Durán and Jongsma advocate for a more transparent design and implementation of black box algorithms, providing a series of recommendations to mitigate the epistemic and ethical challenges behind their use in medical practice. In this response, I argue that a peculiar form of black box algorithm transparency, called design publicity, may efficiently implement these recommendations. Design publicity encodes epistemic, that is, reliability indicators, and ethical recommendations for black box algorithms by means of four subtypes of transparency. These target the values and goals, their translation into design requirements, the performance and consistency of the algorithm altogether. I discuss design publicity applying it to a use case focused on the automated classification of skin lesions from medical images. (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)

We propose a formal framework for interpretable machine learning. Combining elements from statistical learning, causal interventionism, and decision theory, we design an idealisedexplanation gamein which players collaborate to find the best explanation(s) for a given algorithmic prediction. Through an iterative procedure of questions and answers, the players establish a three-dimensional Pareto frontier that describes the optimal trade-offs between explanatory accuracy, simplicity, and relevance. Multiple rounds are played at different levels of abstraction, allowing the players to explore overlapping causal patterns of (...) variable granularity and scope. We characterise the conditions under which such a game is almost surely guaranteed to converge on a (conditionally) optimal explanation surface in polynomial time, and highlight obstacles that will tend to prevent the players from advancing beyond certain explanatory thresholds. The game serves a descriptive and a normative function, establishing a conceptual space in which to analyse and compare existing proposals, as well as design new and improved solutions. (shrink)

The increasing accuracy of algorithms to predict values and preferences raises the possibility that artificial intelligence technology will be able to serve as a surrogate decision-maker for incapacitated patients. Following Camillo Lamanna and Lauren Byrne, we call this technology the autonomy algorithm. Such an algorithm would mine medical research, health records, and social media data to predict patient treatment preferences. The possibility of developing the AA raises the ethical question of whether the AA or a relative ought to serve as (...) surrogate decision-maker in cases where the patient has not issued a medical power of attorney. We argue that in such cases, and against the standard practice of vesting familial surrogates with decision making authority, the AA should have sole decision-making authority. This is because the AA will likely be better at predicting what treatment option the patient would have chosen. It would also be better at avoiding bias and, therefore, choosing in a more patient-centered manner. Furthermore, we argue that these considerations override any moral weight of the patient’s special relationship with their relatives. (shrink)

This paper presents a novel philosophical analysis of the problem of law enforcement’s use of biased face recognition technology in liberal democracies. FRT programs used by law enforcement in identifying crime suspects are substantially more error-prone on facial images depicting darker skin tones and females as compared to facial images depicting Caucasian males. This bias can lead to citizens being wrongfully investigated by police along racial and gender lines. The author develops and defends “A Liberal Argument Against Biased FRT,” which (...) concludes that law enforcement use of biased FRT is inconsistent with the classical liberal requirement that government treat all citizens equally before the law. Two objections to this argument are considered and shown to be unsound. The author concludes by suggesting that equality before the law should be preserved while the problem of machine bias ought to be resolved before FRT and other types of artificial intelligence are deployed by governments in liberal democracies. (shrink)

This article presents Epistemic Spatialization as a new framework for investigating the interconnected patterns of biases when identifying objects with convolutional neural networks (convnets). It draws upon Foucault’s notion of spatialized knowledge to guide its method of enquiry. We argue that decisions involved in the creation of algorithms, alongside the labeling, ordering, presentation, and commercial prioritization of objects, together create a distorted “nomination of the visible”: they harden the visibility of some objects, make other objects excessively visible, and consign yet (...) others to permanent or haphazard invisibility. Our approach differs from those who focus on high-stakes misidentifications, such as errors tied to structural racism. Examining the far more dominant series of low-stakes mistakes shows the scope of errors, destabilizing the goal of image content identification with considerable societal impact. We explore these issues by closely examining the demonstration video of a popular convnet. This examination reveals an interlocking series of biases undermining the content identification process. The picture we paint is crucial for a better understanding of the errors that result as these convnets become further embedded in everyday products. The framework is valuable for critical work on computer vision, AI studies, and large-scale visual analysis. (shrink)

Early and accurate diagnosis of glaucoma is critical for avoiding human vision deterioration and preventing blindness. A deep-neural-network model has been developed for the diagnosis of glaucoma based on Heidelberg retina tomography, called “Seeking Common Features and Reserving Differences Net” to make full use of the HRT data. In this work, the proposed SCRD-Net model achieved an area under the curve of 94.0%. For the two HRT image modalities, the model sensitivities were 91.2% and 78.3% at specificities of 0.85 and (...) 0.95, respectively. These results demonstrate a significant improvement over earlier results. In addition, we visualized the network outputs to develop an interpretation of the learned mechanism for discriminating glaucoma and normal images. Thus, the SCRD-Net can be an effective diagnostic indicator of glaucoma during clinical screening. To facilitate SCRD-Net utilization by the scientific community, the code implementation will be made publicly available. (shrink)