The ethical review of research using methods of artificial intelligence and machine learning in health care contexts has become an important challenge for Research Ethics Boards (also refer...

Applications of artificial intelligence/machine learning in health care are dynamic and rapidly growing. One strategy for anticipating and addressing ethical challenges related to AI/ml for health care is patient and public involvement in the design of those technologies – often referred to as ‘co-design’. Co-design has a diverse intellectual and practical history, however, and has been conceptualized in many different ways. Moreover, AI/ml introduces challenges to co-design that are often underappreciated. Informed by perspectives from critical data studies and critical digital (...) health studies, we review the research literature on involvement in health care, and involvement in design, and examine the extent to which co-design as commonly conceptualized is capable of addressing the range of normative issues raised by AI/ml for health care. We suggest that AI/ml technologies have amplified and modified existing challenges related to patient and public involvement, and created entirely new challenges. We outline three pitfalls associated with co-design for ethical AI/ml for health care and conclude with suggestions for addressing these practical and conceptual challenges. (shrink)

Blackmail raises a pair of parallel legal and moral problems, sometimes referred to as the "paradox of blackmail". It is sometimes legal and morally permissible to ask someone for money, or to threaten to release harmful information about them, while it is illegal and morally impermissible to do these actions jointly. I address the moral version of this paradox by bringing instances of blackmail under a general account of wrongful coercion. According to this account, and contrary to the appearances which (...) give rise to the paradox, threatening the release of harmful information to constrain another's actions is almost never morally impermissible unless it is likewise impermissible to carry out one's threat. To defend this claim I identify a special wrong that arises in the paradoxical cases of threatened information release. The account also resolves a number of other puzzles about blackmail—for example, why profiting from a threat in blackmail can sometimes be impermissible, even though accepting identical payment from an independent offer to retain the same information can be permissible. (shrink)

Background Our objective was to evaluate whether the description of a machine learning (ML) app or brain imaging technology to predict the onset of schizophrenia or alcohol use disorder (AUD) influences healthcare professionals’ judgments of stigma, empathy, and compassion.Methods We randomized healthcare professionals (N = 310) to one vignette about a person whose clinician seeks to predict schizophrenia or an AUD, using a ML app, brain imaging, or a psychosocial assessment. Participants used scales to measure their judgments of stigma, empathy, (...) and compassion.Results Participants randomized to the ML vignette endorsed less anger and more fear relative to the psychosocial vignette, and the brain imaging vignette elicited higher pity ratings. The brain imaging and ML vignettes evoked lower personal responsibility judgments compared to the psychosocial vignette. Physicians and nurses reported less empathy than clinical psychologists.Conclusions The use of predictive technologies may reinforce essentialist views about mental health and substance use that may increase specific aspects of stigma and reduce others. (shrink)

De se exceptionalism is the view, notably championed by Perry (1979) and Lewis (1979), that our characteristically 'first-personal' ways of thinking about ourselves present unique challenges to standard views of propositional attitudes like belief. Though the view has won many adherents, it has recently come under a barrage of deserved criticism. A key claim of detractors is that classic examples used to motivate de se exceptionalism from de se ignorance or misidentification are nothing more than familiar Frege-puzzles, which raise no (...) issues exclusive to self-directed thought. After reviewing how this simple objection has substantial force against the classic defenses of exceptionalism, I provide new arguments based on cases of pure de se ignorance that avoid the criticism. Afterward, I revisit Lewis's defense of exceptionalism, diagnosing how he arrived at roughly the right conclusions on the basis of an argument that begged critical questions and obscured the representationally unique features of the de se. (shrink)

Truth-value gaps have received little attention from a foundational perspective, a fact which has rightfully opened up gap theories to charges of vacuousness. This paper develops an account of the foundations of gap-like behavior which has some hope of avoiding such charges. I begin by reviewing and sharpening a powerful argument of Dummett’s to constrain the options that gap theorists have to make sense of their views. I then show that within these strictures, we can give an account of gaps (...) by drawing on elements of a broadly Stalnakerian framework for assertion and using gaps to track an amalgamation of assertoric effects. The discussion reveals that we may need special resources in our theories of assertion to posit gaps, that gaps may be unusable in characterizing the structure of mental states, and that gaps may have heterogeneous linguistic sources that result in equally heterogeneous projective and inferential behavior. (shrink)

A variety of approaches have appeared in academic literature and in design practice representing “ethics-first” methods. These approaches typically focus on clarifying the normative dimensions of design, or outlining strategies for explicitly incorporating values into design. While this body of literature has developed considerably over the last 20 years, two themes central to the endeavour of ethics and values in design (E + VID) have yet to be systematically discussed in relation to each other: (a) designer agency, and (b) the (...) strength of normative claims informing the design process. To address this gap, we undertook a structured review of leading E + VID approaches and critiques, and classified them according to their positions on normative strength, and views regarding designer agency. We identified 18 distinct approaches and 13 critiques that met the inclusion criteria for our review. Included papers were distributed across the spectrum of views regarding normative strength, and we found that no approaches and only one critique represented a view characteristic of “low” designer agency. We suggest that the absence of “low” designer agency approaches results in the neglect of crucial influences on design as targets of intervention by designers. We conclude with suggestions for future research that might illuminate strategies to achieve ethical design in information mature societies, and argue that without attending to the tensions raised by balancing normatively “strong” visions of the future with limitations imposed on designer agency in corporate-driven design settings, “meaningful” ethical design will continue to encounter challenges in practice. (shrink)

I argue that on very weak assumptions about truth (in particular, that there are coherent norms governing the use of "true"), there is a proposition absolutely inexpressible with conventional language, or something very close. I argue for this claim "constructively": I use a variant of the Berry Paradox to reveal a particular thought for my readership to entertain that very strongly resists conventional expression. I gauge the severity of this expressive limitation within a taxonomy of expressive failures, and argue that (...) despite its strength there is nothing incoherent about admitting its existence. The argument forms part of a project of clarifying precisely what trade-offs are required to secure the kinds of expressive power truth theorists typically want, in the process showing that the admission of very strong expressive limitations may ultimately prove to be the lesser of two evils. (shrink)

The Sleeping Beauty puzzle has dramatized the divisive question of how de se beliefs should be integrated into formal theories of rational belief change. In this paper, I look ahead to a related question: how should de se beliefs be integrated into formal theories of rational choice? I argue that standard decision theoretic frameworks fail in special cases of de se uncertainty, like Sleeping Beauty. The nature of the failure reveals that sometimes rational choices are determined independently of one’s credences (...) in the kinds of ‘narrow’ de se propositions that Sleepy Beauty has set in relief. Consequently, in addition to pinpointing a failure of standard decision theoretic frameworks, this result casts doubt on a large class of strategies for determining principles for the rationally updating de se beliefs in cases like Sleeping Beauty, and also calls into question the importance of making such a determination at all. (shrink)

I argue for two theses about semantically anomalous utterances (more commonly called "category mistakes") like "sequestered slaps reel evergreen rights". First, semantic anomaly generates a unique form of semantically enforced quantifier domain restriction. Second, the best explanation for why anomaly interacts with quantifiers in this way is that anomalous utterances are truth-valueless. After arguing for these points, I trace out two consequences these theses have in semantics and logic. First, I argue they motivate a trivalent semantics on which truth-valueless material (...) has an unsual positive role to play in the compositional semantics of truth-evaluable utterances. Second, I argue that the interaction of anomaly with quantifiers generates a unique form of classical inference failure, and also provides special motivations for reconceptualizing our logical consequence relations. (shrink)

The inclusion of members of structurally marginalized communities in data-intensive innovation initiatives, such as precision medicine projects, is an urgent contemporary issue. On the one hand, th...

Cumming (2008) uses a puzzle about belief ascription to argue against a Millian semantics, and in favor of a semantics on which names are assigned denotations relative to a shiftable variable assignment. I use Cumming's puzzle to showcase the virtues of a rival, broadly Stalnakerian, treatment of attitude ascriptions that safeguards Millianism. I begin by arguing that Cumming's solution seems unable to account for substitutivity data that helps constitute the very puzzle he uses to motivate his account. Once the substitutivity (...) data is acknowledged, the puzzle actually seems to strengthen, rather than weaken, the Millian position. I then argue that the real force of the puzzle is to place special constraints on our accounts of de re belief ascription. To help bring out the nature of those constraints, I connect aspects of Cumming's puzzle to more familiar problems about belief ascription raised by Quine. I then defend three theses making up a skeletal account of de re belief ascription that promises to cope naturally with the variants of Cumming's puzzle, and integrate them into a broader theory of de re intensionality. (shrink)

In Category Mistakes, Ofra Magidor provides two arguments against the view that category mistakes express propositions that are not truth-evaluable at some world. I argue that her first, Williamsonian argument against this view begs the question, and that her second argument rests on a misleading conception of how semantic defect results in infelicity judgments. I conclude by conceding that she is still correct to stress that the view she opposes face noteworthy foundational and empirical challenges.

The goal of this book is to develop a new approach to reading the rule-following sections guided by a simple idea. The simple idea is that Wittgenstein's remarks on rule-following are split between two distinct but complementary projects. The projects are marked not only by different guiding questions, but different presuppositions and methodologies. There is of course precedent for reading the rule-following remarks as comprising two parts. For example, there is the reading of (S. Kripke 1982) on which Wittgenstein first (...) presents a skeptical challenge before turning to its skeptical solution. And there are readings like (Minar 2011) and (Goldfarb 2012) that likewise treat the sections as roughly divided into two components or "chapters." But even these commentators do not explicitly associate their components with different questions and associated methodologies, and so fall short of attributing to Wittgenstein's text anything quite like the thoroughgoing bipartite structure I will be proposing here. (shrink)

Background The ethical governance of Artificial Intelligence (AI) in health care and public health continues to be an urgent issue for attention in policy, research, and practice. In this paper we report on central themes related to challenges and strategies for promoting ethics in research involving AI in global health, arising from the Global Forum on Bioethics in Research (GFBR), held in Cape Town, South Africa in November 2022. Methods The GFBR is an annual meeting organized by the World Health (...) Organization and supported by the Wellcome Trust, the US National Institutes of Health, the UK Medical Research Council (MRC) and the South African MRC. The forum aims to bring together ethicists, researchers, policymakers, research ethics committee members and other actors to engage with challenges and opportunities specifically related to research ethics. In 2022 the focus of the GFBR was “Ethics of AI in Global Health Research”. The forum consisted of 6 case study presentations, 16 governance presentations, and a series of small group and large group discussions. A total of 87 participants attended the forum from 31 countries around the world, representing disciplines of bioethics, AI, health policy, health professional practice, research funding, and bioinformatics. In this paper, we highlight central insights arising from GFBR 2022. Results We describe the significance of four thematic insights arising from the forum: (1) Appropriateness of building AI, (2) Transferability of AI systems, (3) Accountability for AI decision-making and outcomes, and (4) Individual consent. We then describe eight recommendations for governance leaders to enhance the ethical governance of AI in global health research, addressing issues such as AI impact assessments, environmental values, and fair partnerships. Conclusions The 2022 Global Forum on Bioethics in Research illustrated several innovations in ethical governance of AI for global health research, as well as several areas in need of urgent attention internationally. This summary is intended to inform international and domestic efforts to strengthen research ethics and support the evolution of governance leadership to meet the demands of AI in global health research. (shrink)