Integrating large language models (LLMs) like GPT-4 into medical ethics is a novel concept, and understanding the effectiveness of these models in aiding ethicists with decision-making can have significant implications for the healthcare sector. Thus, the objective of this study was to evaluate the performance of GPT-4 in responding to complex medical ethical vignettes and to gauge its utility and limitations for aiding medical ethicists. Using a mixed-methods, cross-sectional survey approach, a panel of six ethicists assessed LLM-generated responses to eight (...) ethical vignettes.The main outcomes measured were relevance, reasoning, depth, technical and non-technical clarity, as well as acceptability of GPT-4’s responses. The readability of the responses was also assessed. Of the six metrics evaluating the effectiveness of GPT-4’s responses, the overall mean score was 4.1/5. GPT-4 was rated highest in providing technical (4.7/5) and non-technical clarity (4.4/5), whereas the lowest rated metrics were depth (3.8/5) and acceptability (3.8/5). There was poor-to-moderate inter-rater reliability characterised by an intraclass coefficient of 0.54 (95% CI: 0.30 to 0.71). Based on panellist feedback, GPT-4 was able to identify and articulate key ethical issues but struggled to appreciate the nuanced aspects of ethical dilemmas and misapplied certain moral principles.This study reveals limitations in the ability of GPT-4 to appreciate the depth and nuanced acceptability of real-world ethical dilemmas, particularly those that require a thorough understanding of relational complexities and context-specific values. Ongoing evaluation of LLM capabilities within medical ethics remains paramount, and further refinement is needed before it can be used effectively in clinical settings. (shrink)

Current advances in research, development and application of artificial intelligence systems have yielded a far-reaching discourse on AI ethics. In consequence, a number of ethics guidelines have been released in recent years. These guidelines comprise normative principles and recommendations aimed to harness the “disruptive” potentials of new AI technologies. Designed as a semi-systematic evaluation, this paper analyzes and compares 22 guidelines, highlighting overlaps but also omissions. As a result, I give a detailed overview of the field of AI ethics. Finally, (...) I also examine to what extent the respective ethical principles and values are implemented in the practice of research, development and application of AI systems—and how the effectiveness in the demands of AI ethics can be improved. (shrink)

The fact that real-world decisions made by artificial intelligences (AI) are often ethically loaded has led a number of authorities to advocate the development of “moral machines”. I argue that the project of building “ethics” “into” machines presupposes a flawed understanding of the nature of ethics. Drawing on the work of the Australian philosopher, Raimond Gaita, I argue that ethical dilemmas are problems for particular people and not (just) problems for everyone who faces a similar situation. Moreover, the force of (...) an ethical claim depends in part on the life history of the person who is making it. For both these reasons, machines could at best be engineered to provide a shallow simulacrum of ethics, which would have limited utility in confronting the ethical and policy dilemmas associated with AI. (shrink)

Questions of what a self-driving car ought to do if it encounters a situation analogous to the ‘trolley problem’ have dominated recent discussion of the ethics of self-driving cars. This paper argues that this interest is misplaced. If a trolley-style dilemma situation actually occurs, given the limits on what information will be available to the car, the dynamics of braking and tyre traction determine that, irrespective of outcome, it is always least risky for the car to brake in a straight (...) line rather than swerve. (shrink)

Developing and implementing artificial intelligence (AI) systems in an ethical manner faces several challenges specific to the kind of technology at hand, including ensuring that decision-making systems making use of machine learning are just, fair, and intelligible, and are aligned with our human values. Given that values vary across cultures, an additional ethical challenge is to ensure that these AI systems are not developed according to some unquestioned but questionable assumption of universal norms but are in fact compatible with the (...) societies in which they operate. This is particularly pertinent for AI research and implementation across Africa, a ground where AI systems are and will be used but also a place with a history of imposition of outside values. In this paper, we thus critically examine one proposal for ensuring that decision-making systems are just, fair, and intelligible—that we adopt a principle of explicability to generate specific recommendations—to assess whether the principle should be adopted in an African research context. We argue that a principle of explicability not only can contribute to responsible and thoughtful development of AI that is sensitive to African interests and values, but can also advance tackling some of the computational challenges in machine learning research. In this way, the motivation for ensuring that a machine learning-based system is just, fair, and intelligible is not only to meet ethical requirements, but also to make effective progress in the field itself. (shrink)

Artificial intelligence plays an important role in current discussions on information and communication technologies and new modes of algorithmic governance. It is an unavoidable dimension of what social mediations and modes of reproduction of our information societies will be in the future. While several works in artificial intelligence ethics address ethical issues specific to certain areas of expertise, these ethical reflections often remain confined to narrow areas of application, without considering the global ethical issues in which they are embedded. We, (...) therefore, propose to clarify the main approaches to AIE, their philosophical assumptions and the specific characteristics of each one of them, to identify the most promising approach to develop an ethical reflection on the deployment of AI in our societies, which is the one based on information ethics as proposed by Luciano Floridi. We will identify the most important features of that approach to highlight areas that need further investigation. (shrink)

This paper presents an experimental approach for engaging undergraduate STEM students in anticipatory ethical reasoning, or ethical reasoning applied to the analysis of potential mid- to long-term implications and outcomes of technological innovation. The authors implemented two variations of an approach that integrates three key components—scenario analysis, design fiction, and ethical frameworks—into five sections of an introductory course on the social contexts of science and technology that is required of STEM majors. The authors dub this approach Creative Anticipatory Ethical Reasoning, (...) or CAER. Scenario analysis is a strategy emerging from business consulting for grounded analysis of plausible future trajectories to inform planning. Design fiction is a creative hands-on activity that blends science fiction and design prototyping to facilitate critical thinking with respect to the societal dimensions of a plausible future technology. The authors present the following findings: in each of the variations, students demonstrated significant engagement with CAER and a substantive shift in their conception of what constitutes responsible innovation and ethical conduct in science and technology. Specifically, their integration of ethical reasoning with stakeholder perspectives and scenario analysis reframed technologies, from unproblematic solutions for societal problems to socially-embedded forms of life that might diverge from designers’ intentions. This suggests that CAER could be a useful pedagogical intervention for expanding students’ ethical engagement to consider the potential unintended consequences of technological innovation. (shrink)

This theoretical paper argues for prioritarianism as an ethical underpinning for digital health in contexts of extreme disadvantage. In support of this claim, the paper develops three prioritarian principles for making ethical decisions for digital health programme design, grounded in the normative position that the greater the need, the stronger the moral claim. The principles are positioned as an alternative view to the prevailing utilitarian approach to digital health, which the paper argues is not sufficient to address the needs of (...) the worst off. As researchers of digital health, we must ensure that the most globally marginalised are not overlooked by overtly technocentric implementation practices. Consequently, the paper concludes by advocating for use of the three principles to support stronger critical reflection on the ethics involved in the design and implementation of digital health programmes. (shrink)

The literature on ethics and user attitudes towards AVs discusses user concerns in relation to automation; however, we show that there are additional relevant issues at stake. To assess adolescents’ attitudes regarding the ‘car of the future’ as presented by car manufacturers, we conducted two studies with over 400 participants altogether. We used a mixed methods approach in which we combined qualitative and quantitative methods. In the first study, our respondents appeared to be more concerned about other aspects of AVs (...) than automation. Instead, their most commonly raised concerns were the extensive use of AI, recommender systems, and related issues of autonomy, invasiveness and personal privacy. The second study confirmed that several AV impacts were negatively perceived. The responses were, however, ambivalent. This confirms previous research on AV attitudes. On one hand, the AV features were perceived as useful, while on the other hand, their impacts were negatively assessed. We followed theoretical insights form futures studies and responsible research and innovation, which helped to identify that there are additional user concerns than what has been previously discussed in the literature on public attitudes and ethics of AVs, as well what has been envisioned by car manufactures. (shrink)

This article introduces readers to the special issue on Selected Issues in the Ethics of Artificial Intelligence. In this paper, I make a case for a wider outlook on the ethics of AI. So far, much of the engagements with the subject have come from Euro-American scholars with obvious influences from Western epistemic traditions. I demonstrate that socio-cultural features influence our conceptions of ethics and in this case the ethics of AI. The goal of this special issue is to entertain (...) more diverse views, particularly those from Africa; it brings together six articles addressing pertinent issues in the ethics of AI. These articles address topics around artificial moral agency, patiency, personhood, social robotics, and the principle of explicability. These works offer unique contributions for and from an African perspective. I contend that a wider engagement with the ethics of AI is worthwhile as we anticipate a global deployment of artificial intelligence systems. (shrink)

This article addresses a dilemma about autonomous vehicles: how to respond to trade-off scenarios in which all possible responses involve the loss of life but there is a choice about whose life or lives are lost. I consider four options: kill fewer people, protect passengers, equal concern for survival, and recognize everyone’s interests. I solve this dilemma via what I call the new trolley problem, which seeks a rationale for the intuition that it is unethical to kill a smaller number (...) of people to avoid killing a greater number of people based on numbers alone. I argue that killing a smaller number of people to avoid killing a greater number of people based on numbers alone is unethical because it disrespects the humanity of the individuals in the smaller-numbered group. I defend the recognize-everyone’s-interests algorithm, which will probably kill fewer people but will not do so based on numbers alone. (shrink)

How should automated vehicles react in emergency circumstances? Most research projects and scientific literature deal with this question from a moral perspective. In particular, it is customary to treat emergencies involving AVs as instances of moral dilemmas and to use the trolley problem as a framework to address such alleged dilemmas. Some critics have pointed out some shortcomings of this strategy and have urged to focus on mundane traffic situations instead of trolley cases involving AVs. Besides, these authors rightly point (...) out the political nature of the most interesting debates involving AVs. However, in our view, they do not offer an adequate account of the distinction between ethics and politics and still see their proposals as contributions to the ethics of AVs. We argue that many of the interesting questions about how AVs should behave, both in emergency and other situations, are of political, and not moral, nature. This view is based on a conception of politics and political normativity that we have developed elsewhere and that we call “political minimalism.” Additionally, we show that this proposed perspective has significant consequences for the design, management, and regulation of transport systems. (shrink)

Artificial intelligence is becoming increasingly entwined with our daily lives: AIs work as assistants through our phones, control our vehicles, and navigate our vacuums. As these objects become more complex and work within our societies in ways that affect our well-being, there is a growing demand for machine ethics—we want a guarantee that the various automata in our lives will behave in a way that minimizes the amount of harm they create. Though many technologies exist as moral artifacts, the development (...) of a truly ethical AI system is highly contentious; theorists have proposed and critiqued countless possibilities for programming these agents to become ethical. Many of these arguments, however, presuppose the possibility that an artificially intelligent system can actually be ethical. In this essay, I will explore a potential path to AI ethics by considering the role of imagination in the deliberative process via the work of John Dewey and his interpreters, showcasing one form of reinforcement learning that mimics imaginative deliberation. With these components in place, I contend that such an artificial agent is capable of something very near ethical behavior—close enough that we may consider it so. (shrink)

Empirical research in bioethics has developed rapidly over the past decade, but has largely eschewed the use of technology-driven methodologies. We propose “design bioethics” as an area of conjoined theoretical and methodological innovation in the field, working across bioethics, health sciences and human-centred technological design. We demonstrate the potential of digital tools, particularly purpose-built digital games, to align with theoretical frameworks in bioethics for empirical research, integrating context, narrative and embodiment in moral decision-making. Purpose-built digital tools can engender situated engagement (...) with bioethical questions; can achieve such engagement at scale; and can access groups traditionally under-represented in bioethics research and theory. If developed and used with appropriate rigor, tools motivated by “design bioethics” could offer unique insights into new and familiar normative and empirical issues in the field. (shrink)

As AI Systems become increasingly autonomous, they are expected to engage in decision-making processes that have moral implications. In this research we integrate theoretical and empirical lines of thought to address the matters of moral reasoning and moral uncertainty in AI Systems. We reconceptualize the metanormative framework for decision-making under moral uncertainty and we operationalize it through a latent class choice model. The core idea being that moral heterogeneity in society can be codified in terms of a small number of (...) classes with distinct moral preferences and that this codification can be used to express moral uncertainty of an AI. Choice analysis allows for the identification of classes and their moral preferences based on observed choice data. Our reformulation of the metanormative framework is theory-rooted and practical in the sense that it avoids runtime issues in real time applications. To illustrate our approach we conceptualize a society in which AI Systems are in charge of making policy choices. While one of the systems uses a baseline morally certain model, the other uses a morally uncertain model. We highlight cases in which the AI Systems disagree about the policy to be chosen, thus illustrating the need to capture moral uncertainty in AI systems. (shrink)

Dilemma situations involving the choice of which human life to save in the case of unavoidable accidents are expected to arise only rarely in the context of autonomous vehicles. Nonetheless, the scientific community has devoted significant attention to finding appropriate and acceptable automated decisions in the event that AVs or drivers of AVs were indeed to face such situations. Awad and colleagues, in their now famous paper “The Moral Machine Experiment”, used a “multilingual online ‘serious game’ for collecting large-scale data (...) on how citizens would want AVs to solve moral dilemmas in the context of unavoidable accidents.” Awad and colleagues undoubtedly collected an impressive and philosophically useful data set of armchair intuitions. However, we argue that applying their findings to the development of “global, socially acceptable principles for machine learning” would violate basic tenets of human rights law and fundamental principles of human dignity. To make its arguments, our paper cites principles of tort law, relevant case law, provisions from the Universal Declaration of Human Rights, and rules from the German Ethics Code for Autonomous and Connected Driving. (shrink)

Achieving the global benefits of artificial intelligence (AI) will require international cooperation on many areas of governance and ethical standards, while allowing for diverse cultural perspectives and priorities. There are many barriers to achieving this at present, including mistrust between cultures, and more practical challenges of coordinating across different locations. This paper focuses particularly on barriers to cooperation between Europe and North America on the one hand and East Asia on the other, as regions which currently have an outsized impact (...) on the development of AI ethics and governance. We suggest that there is reason to be optimistic about achieving greater cross-cultural cooperation on AI ethics and governance. We argue that misunderstandings between cultures and regions play a more important role in undermining cross-cultural trust, relative to fundamental disagreements, than is often supposed. Even where fundamental differences exist, these may not necessarily prevent productive cross-cultural cooperation, for two reasons: (1) cooperation does not require achieving agreement on principles and standards for all areas of AI; and (2) it is sometimes possible to reach agreement on practical issues despite disagreement on more abstract values or principles. We believe that academia has a key role to play in promoting cross-cultural cooperation on AI ethics and governance, by building greater mutual understanding, and clarifying where different forms of agreement will be both necessary and possible. We make a number of recommendations for practical steps and initiatives, including translation and multilingual publication of key documents, researcher exchange programmes, and development of research agendas on cross-cultural topics. (shrink)

This paper takes stock of all the various factors that cause the design-time opacity of autonomous systems behaviour. The factors include embodiment effects, design-time knowledge gap, human factors, emergent behaviour and tacit knowledge. This situation is contrasted with the usual representation of moral dilemmas that assume perfect information. Since perfect information is not achievable, the traditional moral dilemma representations are not valid and the whole problem of ethical autonomous systems design proves to be way more empirical than previously understood.

This paper considers possible future legislation that requires the exclusive use of autonomous cars. The author develops and defends a ‘Liberal Argument Against Mandated Autonomous Cars’, which argues that such a law would be incompatible with classical liberalism, provided that the following condition holds: In the event where the car must ‘choose’ between running over a young person or an old person, or both, autonomous cars are programmed to respond by running over old people in order to save young people. (...) Such a law requiring the use of these cars, provided that these assumptions hold, would violate the important value in classical liberalism that all individuals ought to be treated equally before the law. The paper concludes by arguing that alternative ways of dealing with this problem come with their own set of unpalatable problems. (shrink)

This paper considers possible future legislation that requires the exclusive use of autonomous cars. The author develops and defends a ‘Liberal Argument Against Mandated Autonomous Cars’, which argues that such a law would be incompatible with classical liberalism, provided that the following condition holds: In the event where the car must ‘choose’ between running over a young person or an old person, or both, autonomous cars are programmed to respond by running over old people in order to save young people. (...) Such a law requiring the use of these cars, provided that these assumptions hold, would violate the important value in classical liberalism that all individuals ought to be treated equally before the law. The paper concludes by arguing that alternative ways of dealing with this problem come with their own set of unpalatable problems. (shrink)

This paper looks at philosophical questions that arise in the context of AI alignment. It defends three propositions. First, normative and technical aspects of the AI alignment problem are interrelated, creating space for productive engagement between people working in both domains. Second, it is important to be clear about the goal of alignment. There are significant differences between AI that aligns with instructions, intentions, revealed preferences, ideal preferences, interests and values. A principle-based approach to AI alignment, which combines these elements (...) in a systematic way, has considerable advantages in this context. Third, the central challenge for theorists is not to identify ‘true’ moral principles for AI; rather, it is to identify fair principles for alignment that receive reflective endorsement despite widespread variation in people’s moral beliefs. The final part of the paper explores three ways in which fair principles for AI alignment could potentially be identified. (shrink)

The ethics of autonomous vehicles has received a great amount of attention in recent years, specifically in regard to their decisional policies in accident situations in which human harm is a likely consequence. Starting from the assumption that human harm is unavoidable, many authors have developed differing accounts of what morality requires in these situations. In this article, a strategy for AV decision-making is proposed, the Ethical Valence Theory, which paints AV decision-making as a type of claim mitigation: different road (...) users hold different moral claims on the vehicle’s behavior, and the vehicle must mitigate these claims as it makes decisions about its environment. Using the context of autonomous vehicles, the harm produced by an action and the uncertainties connected to it are quantified and accounted for through deliberation, resulting in an ethical implementation coherent with reality. The goal of this approach is not to define how moral theory requires vehicles to behave, but rather to provide a computational approach that is flexible enough to accommodate a number of ‘moral positions’ concerning what morality demands and what road users may expect, offering an evaluation tool for the social acceptability of an autonomous vehicle’s ethical decision making. (shrink)

Autonomous vehicles —and accidents they are involved in—attest to the urgent need to consider the ethics of artificial intelligence. The question dominating the discussion so far has been whether we want AVs to behave in a ‘selfish’ or utilitarian manner. Rather than considering modeling self-driving cars on a single moral system like utilitarianism, one possible way to approach programming for AI would be to reflect recent work in neuroethics. The agent–deed–consequence model :3–20, 2014a, Behav Brain Sci 37:487–488, 2014b) provides a (...) promising descriptive and normative account while also lending itself well to implementation in AI. The ADC model explains moral judgments by breaking them down into positive or negative intuitive evaluations of the agent, deed, and consequence in any given situation. These intuitive evaluations combine to produce a positive or negative judgment of moral acceptability. For example, the overall judgment of moral acceptability in a situation in which someone committed a deed that is judged as negative would be mitigated if the agent had good intentions and the action had a good consequence. This explains the considerable flexibility and stability of human moral judgment that has yet to be replicated in AI. This paper examines the advantages and disadvantages of implementing the ADC model and how the model could inform future work on ethics of AI in general. (shrink)

Machines are increasingly involved in decisions with ethical implications, which require ethical explanations. Current machine learning algorithms are ethically inscrutable, but not in a way very different from human behavior. This article looks at the role of rationality and reasoning in traditional ethical thought and in artificial intelligence, emphasizing the need for some explainability of actions. It then explores Neil Lawrence's embodiment factor as an insightful way of looking at the differences between human and machine intelligence, connecting it to the (...) theological understanding of embodiment, relationality, and personhood. Finally, it proposes the notion of artificial moral orthoses, which could provide ethical explanations for both artificial and human agents, as a more promising unifying approach to human and machine ethics. (shrink)