Real engines of the artificial intelligence revolution, machine learning models, and algorithms are embedded nowadays in many services and products around us. As a society, we argue it is now necessary to transition into a phronetic paradigm focused on the ethical dilemmas stemming from the conception and application of AIs to define actionable recommendations as well as normative solutions. However, both academic research and society-driven initiatives are still quite far from clearly defining a solid program of study and intervention. In (...) this contribution, we will focus on selected ethical investigations around AI by proposing an incremental model of trust that can be applied to both human-human and human-AI interactions. Starting with a quick overview of the existing accounts of trust, with special attention to Taddeo’s concept of “e-trust,” we will discuss all the components of the proposed model and the reasons to trust in human-AI interactions in an example of relevance for business organizations. We end this contribution with an analysis of the epistemic and pragmatic reasons of trust in human-AI interactions and with a discussion of kinds of normativity in trustworthiness of AIs. (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)

This edition of the journal includes, with an introduction and three commentaries, a recent joint statement from the British Medical Association, the Resuscitation Council (UK) and the Royal College of Nursing, on decisions relating to cardiopulmonary resuscitation (CPR).1 The statement was produced in response both to a professional need to decide when attempting CPR is and is not ethically appropriate, or indeed lawful (especially in the light of incorporation of the European Convention on Human Rights into UK law), and also (...) to public concern that such decisions have not always been made in concordance with the wishes of patients or people close to them. The guidelines it sets out reflect a broad consensus of current medicolegal views in the UK, include a variety of helpful and practical recommendations, and can be seen, in many respects, as a model of their kind. As a model of actual decision making however, they raise some problems. Model helps us understand the reality they represent by simplifying it, but in the process of simplification, aspects of the reality represented which do not fit in with the model may be left out or downplayed. Three such aspects identified by our commentators concern patient autonomy, quality of life judgments, and the role of families in decision making. In his commentary, Robin Gill argues that while “many of us cherish” the right to be properly informed before decisions about our treatment are made, such information “may become a burden for others when they are at their most vulnerable”. Being told about CPR, with the implication that a “do not attempt resuscitation” (DNAR) order is at least possible, may provoke “angst about death” in these vulnerable patients. The guidelines state that a competent patient's wish not to discuss resuscitation should be respected. But they also state that …. (shrink)

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

Background: Current practice relies on patient-designated and next-of-kin surrogates, in consultation with clinicians, to make treatment decisions for patients who lose the ability to make their own decisions. Yet there is a paucity of data on whether this approach is consistent with patients' preferences regarding who they want to make treatment decisions for them in the event of decisional incapacity. Methods: Self-administered survey of patients at a tertiary care center. Results: Overall, 1169 respondents completed the survey (response rate = 59.8%). (...) Of the 229 respondents who had previously designated a surrogate, 78.2% wanted their surrogate to make treatment decisions in the event of decisional incapacity, whereas 21.8% wanted their doctors to make treatment decisions. Of the 822 respondents who had not designated a surrogate, 66.1% wanted their family to make treatment decisions, whereas 33.9% wanted their doctors to make treatment decisions. The most common explanation provided for why respondents wanted their surrogate or family to make treatment decisions for them in the event of decisional incapacity was the belief that loved ones knew the patient's treatment preferences. Conclusions: Contrary to current practice, 33.9% of respondents who had not designated a surrogate and 21.8% of those who had designated a surrogate indicated that they wanted their doctors to make treatment decisions for them in the event of decisional incapacity. Moreover, many of those who wanted their surrogates or family members to make treatment decisions explained this preference by citing a belief that loved ones knew the patient's treatment preferences. This belief is undermined by prior research that suggests that surrogates and family members frequently are unable to predict which treatments their charges would want. Future research should assess these two concerns with current practice and, where necessary, identify approaches to address them. (shrink)

Do-not-resuscitate (DNR) orders are typically signed by physicians in conjunction with patients or their surrogate decision makers in order to instruct healthcare providers not to perform cardiopulmonary resuscitation (CPR). Both the medical literature and CPR guidelines fail to address when it is appropriate for physicians to sign DNR orders without any knowledge of a patient’s wishes. We explore the ethical issues surrounding instituting a twophysician DNR for a dying patient with multiple comorbidities and no medical record on file, no advance (...) directives, and no surrogate decision maker. Through this case we also highlight the issues of poor prognostication and the reversal of a DNR in such circumstances. (shrink)

Objective:To evaluate the effect of an intervention on the understanding and use of DNR orders by physicians; to assess the impact of understanding the importance of involving competent patients in DNR decisions.Design:Prospective clinical interventional study.Setting:Internal medicine department (70 beds) of the hospital of La Chaux-de-Fonds, Switzerland.Participants:Nine junior physicians in postgraduate training.Intervention:Information on the ethics of DNR and implementation of new DNR orders.Measurements and main results:Accurate understanding, interpretation, and use of DNR orders, especially with respect to the patients’ involvement in the (...) decision were measured. Junior doctors writing DNR orders had an extremely poor understanding of what DNR orders mean. The correct understanding of the definition of a DNR order increased from 31 to 93% (p<0.01) after the intervention and the patients’ involvement went from 17% to 48% (p<0.01). Physicians estimated that 75% of their DNR patients were mentally competent at the time of the decision.Conclusion:An intervention aimed at explaining the ethical principles and the definition of DNR orders improves understanding of them, and their implementation, as well as patient participation. Specific efforts are needed to increase the involvement of mentally competent patients in the decision. (shrink)

Background The ethical principle of justice demands that resources be distributed equally and based on evidence. Guidelines regarding forgoing of CPR are unavailable and there is large variance in the reported rates of attempted CPR in in-hospital cardiac arrest. The main objective of this work was to study whether local culture and physician preferences may affect spur-of-the-moment decisions in unexpected in-hospital cardiac arrest. Methods Cross sectional questionnaire survey conducted among a convenience sample of physicians that likely comprise code team members (...) in their country. The questionnaire included details regarding respondent demographics and training, personal value judgments and preferences as well as professional experience regarding CPR and forgoing of resuscitation. Results Of the 675 questionnaires distributed, 617 were completed and returned. Country of practice and level of knowledge about resuscitation were strongly associated with avoiding CPR performance. Mexican physicians were almost twicemore likely to forgo CPR than their Israeli and Indonesian/Malaysian counterparts [OR1.84, p = 0.038]. Mexican responders also placed greater emphasison personal and patient quality of life. In multivariate analysis, degree of religiosity was most strongly associated with willingness to forgo CPR; orthodox respondents were more than twice more likely to report having forgone CPR for apatient they do not know than secular and observant respondents, regardless of the country of practice [OR 2.12, p = 0.003]. Conclusions In unexpected in-hospital cardiac arrest the decision to perform or withhold CPR may be affected by physician knowledge and local culture as well as personal preferences. Physician CPR training should include information regarding predictors of patient outcome at as well as emphasis on differentiating between patient and personal preferences in an emergency. (shrink)

Objective: To evaluate the effect of an intervention on the understanding and use of DNR orders by physicians; to assess the impact of understanding the importance of involving competent patients in DNR decisions. Design: Prospective clinical interventional study. Setting: Internal medicine department (70 beds) of the hospital of La Chaux-de-Fonds, Switzerland. Participants: Nine junior physicians in postgraduate training. Intervention: Information on the ethics of DNR and implementation of new DNR orders. Measurements and main results: Accurate understanding, interpretation, and use of (...) DNR orders, especially with respect to the patients’ involvement in the decision were measured. Junior doctors writing DNR orders had an extremely poor understanding of what DNR orders mean. The correct understanding of the definition of a DNR order increased from 31 to 93% (p<0.01) after the intervention and the patients’ involvement went from 17% to 48% (p<0.01). Physicians estimated that 75% of their DNR patients were mentally competent at the time of the decision. Conclusion: An intervention aimed at explaining the ethical principles and the definition of DNR orders improves understanding of them, and their implementation, as well as patient participation. Specific efforts are needed to increase the involvement of mentally competent patients in the decision. (shrink)

In contrast to Di Nucci’s characterisation, my argument is not a technoapocalyptic one. The view I put forward is that systems like IBM’s Watson for Oncology create both risks and opportunities from the perspective of shared decision-making. In this response, I address the issues that Di Nucci raises and highlight the importance of bioethicists engaging critically with these developing technologies.

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)