More than a decade of exacting scientific research involving paleontological fragments and ancient DNA has lately produced a series of pronouncements about a purportedly novel population of archaic hominins dubbed “the Denisova.” The science involved in these matters is both technically stunning and, socially, at times a bit reckless. Here I discuss the responsibilities which scientists incur when they make inductively risky pronouncements about the different relative contributions by Denisovans to genomes of members of apparent subpopulations of current humans. This (...) science is sensational: it is science which empirically speculates, to the public delight’s and entertainment, about scintillating topics such as when humans evolved, where we came from, and who else we were having sex with during our early hominin history. An initial characterization of sensational science emerges from my discussion of the case, as well as a diagnosis of an interactive phenomenon termed amplified inductive risk. (shrink)

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)

Under what conditions does machine learning (ML) model opacity inhibit the possibility of explaining and understanding phenomena? In this article, I argue that nonepistemic values give shape to the ML opacity problem even if we keep researcher interests fixed. Treating ML models as an instance of doing model-based science to explain and understand phenomena reveals that there is (i) an external opacity problem, where the presence of inductive risk imposes higher standards on externally validating models, and (ii) an internal opacity (...) problem, where greater inductive risk demands a higher level of transparency regarding the inferences the model makes. (shrink)

In the science and values literature, scholars have shown how science is influenced and shaped by values, often in opposition to the ‘value free’ ideal of science. In this paper, we aim to contribute to the science and values literature by showing that the relation between science and values flows not only from values into scientific practice, but also from (allegedly neutral) science to values themselves. The extant literature in the ‘science and values’ field focuses by and large on reconstructing, (...) post hoc, how values have influenced science; our reconstruction of the case studies, instead, aims to show that scientific concepts and methods _too, because of specific identifiable characteristics,_ can promote some values rather than (or at the expense of) others. We explain this bidirectional relation in analogy to debates on the normativity of technical artifacts and on feminist approaches in science, and we illustrate our claims with cases from the health sciences and machine learning. While our arguments in this paper also draw on post hoc reconstructions, we intend to show where, in the science in the making, we should engage not only with the question whether a practice is value-laden, but also how specific conceptual and methodological choices can influence values down the road. All in all, these considerations expand the ways in which philosophers can contribute to more value-aware scientific practices. (shrink)

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

Optimism about our ability to enhance societal decision‐making by leaning on Machine Learning (ML) for cheap, accurate predictions has palled in recent years, as these ‘cheap’ predictions have come at significant social cost, contributing to systematic harms suffered by already disadvantaged populations. But what precisely goes wrong when ML goes wrong? We argue that, as well as more obvious concerns about the downstream effects of ML‐based decision‐making, there can be moral grounds for the criticism of these predictions themselves. We introduce (...) and defend a theory of predictive justice, according to which differential model performance for systematically disadvantaged groups can be grounds for moral criticism of the model, independently of its downstream effects. As well as helping resolve some urgent disputes around algorithmic fairness, this theory points the way to a novel dimension of epistemic ethics, related to the recently discussed category of doxastic wrong. (shrink)

Science is based onfacts—facts that are systematically gathered by a community of enquirers through detailed observation and experiment. In the twentieth century, however, philosophers of science claimed that the facts that scientists “gather” in this way are shaped by the theories scientists accept, and this seemed to threaten the authority of science. Call this theold worries about science.By contrast, what seemed not to threaten that authority were other factors that shaped the facts that scientists gather—for example, the mere questions scientists (...) pursue. Call this theold nonworries about science.What I suggest is that the old nonworries are turning out to be far more worrisome than the old worries, and I use recent goings-on such as the “Death of Evidence” protests in Canada, the “replication crisis,” and the ongoing feminist critiques of science to illustrate my case. All this raises interesting new questions for philosophers of science to tackle. (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)

I examine the construction and evaluation of machine learning binary classification models. These models are increasingly used for societal applications such as classifying patients into two categories according to the presence or absence of a certain disease like cancer and heart disease. I argue that the construction of ML classification models involves an optimisation process aiming at the minimization of the inductive risk associated with the intended uses of these models. I also argue that the construction of these models is (...) underdetermined by the available data, and that this makes it necessary for ML modellers to make social value judgments in determining the error costs used in ML optimization. I thus suggest that the assessment of the inductive risk with respect to the social values of the intended users is an integral part of the construction and evaluation of ML classification models. I also discuss the implications of this conclusion for the philosophical debate concerning inductive risk. (shrink)

Machine learning algorithms are expected to improve referral decisions. In this article I discuss the legitimacy of deferring referral decisions in primary care to recommendations from such algorithms. The standard justification for introducing algorithmic decision procedures to make referral decisions is that they are more accurate than the available practitioners. The improvement in accuracy will ensure more efficient use of scarce health resources and improve patient care. In this article I introduce a proceduralist framework for discussing the legitimacy of algorithmic (...) referral decisions and I argue that in the context of referral decisions the legitimacy of an algorithmic decision procedure can be fully accounted for in terms of the instrumental values of accuracy and fairness. I end by considering how my discussion of procedural algorithmic legitimacy relates to the debate on algorithmic fairness. (shrink)

Artificial Intelligence and Machine Learning (AI/ML) models introduce unique considerations when determining their epistemic value. Fortunately, existing work on the epistemic features of AI/ML can...

Both scientists and philosophers of science have recently emphasized the importance of promoting transparency in science. For scientists, transparency is a way to promote reproducibility, progress, and trust in research. For philosophers of science, transparency can help address the value-ladenness of scientific research in a responsible way. Nevertheless, the concept of transparency is a complex one. Scientists can be transparent about many different things, for many different reasons, on behalf of many different stakeholders. This paper proposes a taxonomy that clarifies (...) the major dimensions along which approaches to transparency can vary. By doing so, it provides several insights that philosophers and other science studies scholars can pursue. In particular, it helps address common objections to pursuing transparency in science, it clarifies major forms of transparency, and it suggests avenues for further research on this topic. (shrink)

Machine learning (ML) models recently led to major breakthroughs in predictive tasks in the natural sciences. Yet their benefits for the social sciences are less evident, as even high-profile studies on the prediction of life trajectories have shown to be largely unsuccessful – at least when measured in traditional criteria of scientific success. This paper tries to shed light on this remarkable performance gap. Comparing two social science case studies to a paradigm example from the natural sciences, we argue that, (...) in addition to explanation, prediction is an important goal of social science – and we identify constraints that impede pure ML prediction from being successful in that field. As a remedy, we outline elements of an integrative modelling approach that combines explanatory models and predictive ML models. (shrink)

Recent rival attempts in the philosophy of science to put forward a general theory of the properties that all (and only) natural kinds across the sciences possess may have proven to be futile. Instead, I develop a general methodological framework for how to philosophically study kinds. Any kind has to be investigated and articulated together with the human aims that motivate referring to this kind, where different kinds in the same scientific domain can answer to different concrete aims. My core (...) contention is that non-epistemic aims, including environmental, ethical, and political aims, matter as well. This is defended and illustrated based on several examples of kinds, with particular attention to the role of social-political aims: species, race, gender, as well as personality disorders and oppositional defiant disorder as psychiatric kinds. Such non-epistemic aims and values need not always be those personally favoured by scientists, but may have to reflect values that matter to relevant societal stakeholders. Despite the general agenda to study ‘kinds,’ I argue that philosophers should stop using the term ‘natural kinds,’ as this label obscures the relevance of humans interests and the way in which many kinds are based on contingent social processes subject to human responsibility. (shrink)

Philosophical work on science and values has come to engage with the concerns of society and of stakeholders affected by science and policy, leading to socially relevant philosophy of science and socially engaged philosophy of science. This special issue showcases instances of socially relevant philosophy of science, featuring contributions on a diversity of topics by Janet Kourany, Andrew Schroeder, Alison Wylie, Kristen Intemann, Joyce Havstad, Justin Biddle, Kevin Elliott, and Ingo Brigandt.

The prospect of including artificial intelligence in clinical decision-making is an exciting next step for some areas of healthcare. This article provides an analysis of the available kinds of AI systems, focusing on macro-level characteristics. This includes examining the strengths and weaknesses of opaque systems and fully explainable systems. Ultimately, the article argues that “grey box” systems, which include some combination of opacity and transparency, ought to be used in healthcare settings.