Many policies and ethical guidelines recommend developing “trustworthy AI”. We argue that developing morally trustworthy AI is not only unethical, as it promotes trust in an entity that cannot be trustworthy, but it is also unnecessary for optimal calibration. Instead, we show that reliability, exclusive of moral trust, entails the appropriate normative constraints that enable optimal calibration and mitigate the vulnerability that arises in high-stakes hybrid decision-making environments, without also demanding, as moral trust would, the anthropomorphization of AI and thus (...) epistemically dubious behavior. The normative demands of reliability for inter-agential action are argued to be met by an analogue to procedural metacognitive competence (i.e., the ability to evaluate the quality of one’s own informational states to regulate subsequent action). Drawing on recent empirical findings that suggest providing reliability scores (e.g., F1-scores) to human decision-makers improves calibration in the AI system, we argue that reliability scores provide a good index of competence and enable humans to determine how much they wish to rely on the system. (shrink)

When it comes to social robotics (robots that engage human social responses via “eyes” and other facial features, voice-based natural-language interactions, and even evocative movements), ethicists, particularly in European and North American traditions, are divided over whether and why they might be morally considerable. Some argue that moral considerability is based on internal psychological states like consciousness and sentience, and debate about thresholds of such features sufficient for ethical consideration, a move sometimes criticized for being overly dualistic in its framing (...) of mind versus body. Others, meanwhile, focus on the effects of these robots on human beings, arguing that psychological impact alone can qualify an entity for moral status. What both sides overlook is the importance for ordinary moral reasoning of integrating questions about an entity’s “inner life,” and its psychological effect on us. Turning to accounts of relationships in virtue ethics, especially those of the Confucian tradition, we find a more nuanced theory that can provide complex guidance on the moral considerability of social robots, including ethical considerations about whether and how to question this to begin with. (shrink)

Artificial Social Agents (ASAs), which are AI software driven entities programmed with rules and preferences to act autonomously and socially with humans, are increasingly playing roles in society. As their sophistication grows, humans will share greater amounts of personal information, thoughts, and feelings with ASAs, which has significant ethical implications. We conducted a study to investigate what ethical principles are of relative importance when people engage with ASAs and whether there is a relationship between people’s values and the ethical principles (...) they prioritise. The study uses scenarios, embedded with the five AI4People ethical principles (Beneficence, Non-maleficence, Autonomy, Justice, and Explicability), involving ASAs taking on roles traditionally played by humans to understand whether these roles and behaviours (including dialogues) are seen as acceptable or unacceptable. Results from 268 participants reveal the greatest sensitivity to ASA behaviours that relate to Autonomy, Justice, Explicability, and the privacy of their personal data. Models were created using Schwartz’s Refined Values as a possible indicator of how stakeholders discern and prioritise the different AI4People ethical principles when interacting with ASAs. Our findings raise issues around the ethical acceptability of ASAs for nudging and changing behaviour due to participants’ desire for autonomy and their concerns over deceptive ASA behaviours such as pretending to have memories and emotions. (shrink)

In philosophy of mind, zombies are imaginary creatures that are exact physical duplicates of conscious subjects but for whom there is no first-personal experience. Zombies are meant to show that physicalism—the theory that the universe is made up entirely out of physical components—is false. In this paper, I apply the zombie thought experiment to the realm of morality to assess whether moral agency is something independent from sentience. Algorithms, I argue, are a kind of functional moral zombie, such that thinking (...) about the latter can help us better understand and regulate the former. I contend that the main reason why algorithms can be neither autonomous nor accountable is that they lack sentience. Moral zombies and algorithms are incoherent as moral agents because they lack the necessary moral understanding to be morally responsible. To understand what it means to inflict pain on someone, it is necessary to have experiential knowledge of pain. At most, for an algorithm that feels nothing, ‘values’ will be items on a list, possibly prioritised in a certain way according to a number that represents weightiness. But entities that do not feel cannot value, and beings that do not value cannot act for moral reasons. (shrink)

In this paper, I draw on Hannah Arendt’s notion of ‘banality of evil’ to argue that as long as AI systems are designed to follow codes of ethics or particular normative ethical theories chosen by us and programmed in them, they are Eichmanns destined to commit evil. Since intelligence alone is not sufficient for ethical decision making, rather than strive to program AI to determine the right ethical decision based on some ethical theory or criteria, AI should be concerned with (...) avoiding making the wrong decisions, and this requires hardwiring the thinking activity as a prerequisite for decision making. (shrink)

Debates about terrorism and technology often focus on the potential uses of technology by non-state terrorist actors and by states as forms of counterterrorism. Yet, little has been written about how technology shapes how we think about terrorism. In this chapter I argue that technology, and the language we use to talk about technology, constrains and shapes our understanding of the nature, scope, and impact of terrorism, particularly in relation to state terrorism. After exploring the ways in which technology shapes (...) moral thinking, I use two case studies to demonstrate how technology simultaneously hides and enables terroristic forms of state violence: police control technologies and Unmanned Aerial Vehicles (UAVs), or drones. In both these cases, I argue that features of these technologies, combined with a narrative of precision and efficiency, serves to mask the terroristic nature of the violence that these practices inflict and reinforces the moral exclusion of those against whom these technologies are deployed. In conclusion, I propose that identifying acts of terrorism requires a focus on the impact of technologies of violence (whether they are “high tech” or not) on those most affected, regardless of whether users of these technologies conceive of their actions as terroristic. (shrink)

This article tests the proposition that new weapons technology requires Christian ethics to dispense with the just war tradition (JWT) and argues for its development rather than dissolution. Those working in the JWT should be under no illusions, however, that new weapons technologies could (or do already) represent threats to the doing of justice in the theatre of war. These threats include weapons systems that deliver indiscriminate, disproportionate or otherwise unjust outcomes, or that are operated within (quasi-)legal frameworks marked by (...) accountability gaps. The temptation to abrogate (L. abrogare—repeal, evade) responsibility to the machine is also a moral threat to the doing of justice in the theatre of war. (shrink)

This article aims to pay heed to the rational and deliberative dimensions of moral agency within the project of moral enhancement. In this sense, it is presented how the technique of neurofeedback might contribute to the enhancement of moral deliberations and autonomy. Furthermore, this brain-computer interface is thought as a possible element of a Socratic moral assistant interested in improving moral enhancement within a model of full interaction between moral agents and such a moral assistant. This proposal does not embrace (...) the project of moral bioenhancement, centred in the improvement of moral emotions through neuropharmacology. Moreover it rejects the delegation of moral decision-making to moral artificial agents because this would mean that there is continuity between moral human and artificial moral agents. (shrink)

The article examines the challenges involved in the process of developing artificial ethical agents. The process involves the creators or designing professionals, the procedures to develop an ethical agent and the artificial systems. There are two possibilities available to create artificial ethical agents: (a) programming ethical guidance in the artificial Intelligence (AI)-equipped machines and/or (b) allowing AI-equipped machines to learn ethical decision-making by observing humans. However, it is difficult to fulfil these possibilities due to the subjective nature of ethical decision-making. (...) The challenge related to the developers is that they themselves lack training in ethical skills. The creators who develop an artificial ethical agent should be able to foresee the ethical issues and have knowledge about ethical decision-making to improve the ethical use of AI-equipped machines. The suggestion is that the focus should be on training professionals involved in the process of developing these artificial systems in ethics rather than developing artificial ethical agents and thereby attributing ethical agency to it. (shrink)

Monitoring and assistive technologies (MATs) are being used more frequently in healthcare. A central ethical concern is the compatibility of these systems with the moral preferences of their users—an issue especially relevant to participatory approaches within the ethics-by-design debate. However, users’ incapacity to communicate preferences or to participate in design processes, e.g., due to dementia, presents a hurdle for participatory ethics-by-design approaches. In this paper, we explore the question of how the value preferences of users in the field of dementia (...) care can be integrated into AI-based MATs. First, we briefly introduce different ethics-by-design approaches and explain the challenges they face in dementia care. Next, we introduce a novel approach for addressing these challenges. Through a qualitative content analysis of interviews with persons with dementia and their family caregivers (n = 27), we identified multiple ideal–typical value preference profiles. We then applied these profiles in a computer simulation, by which we also introduce the concept of ethical compliance quantification to scale the moral preferences of the interviewees for the simulation. Finally, we discuss study results, the advantages of using computer simulations, and general study limitations before drawing conclusions for further research. The study contributes to the ongoing debate on participatory ethics-by-design by defining on the basis of empirical data ideal–typical value preference frameworks that can be used to guide MAT actions and their outcomes. Going forward, MAT end-users with dementia and other communication-impaired persons could be enabled to choose a value profile that best fits their moral preferences. (shrink)

This paper concerns top-down approaches in machine ethics. It is divided into three main parts. First, I briefly describe top-down design approaches, and in doing so I make clear what those approaches are committed to and what they involve when it comes to training an AI to behave ethically. In the second part, I formulate two underappreciated motivations for endorsing them, one relating to predictability of machine behavior and the other relating to scrutability of machine decision-making. Finally, I present three (...) major worries about such approaches, and I attempt to show that advocates of top-down approaches have some plausible avenues of response. I focus most of my attention on what I call the ‘technical manual objection’ to top-down approaches, inspired by the work of Annas (2004). In short, the idea is that top-down approaches treat ethical decision-making as being merely a matter of following some ethical instructions in the same way that one might follow some set of instructions contained in a technical manual (e.g., computer manual), and this invites sensible skepticism about the ethical wisdom of machines that have been trained on those approaches. I respond by claiming that the objection is successful only if it is understood as targeting machines that have certain kinds of goals, and it should not compel us to totally abandon top-down approaches. Such approaches could still be reasonably employed to design ethical AI that operate in contexts that include fairly noncontroversial answers to ethical questions. In fact, we should prefer top-down approaches when it comes to those types of context, or so I argue, due to the advantages I claim for them. (shrink)

Although some authors claim that deception requires intention, we argue that there can be deception in social robotics, whether or not it is intended. By focusing on the deceived rather than the deceiver, we propose that false beliefs can be created in the absence of intention. Supporting evidence is found in both human and animal examples. Instead of assuming that deception is wrong only when carried out to benefit the deceiver, we propose that deception in social robotics is wrong when (...) it leads to harmful impacts on individuals and society. The appearance and behaviour of a robot can lead to an overestimation of its functionality or to an illusion of sentience or cognition that can promote misplaced trust and inappropriate uses such as care and companionship of the vulnerable. We consider the allocation of responsibility for harmful deception. Finally, we make the suggestion that harmful impacts could be prevented by legislation, and by the development of an assessment framework for sensitive robot applications. (shrink)

This critical review accomplished two main tasks: first, the article provides scope for identifying the most common conceptions of moral competence in the scientific literature, as well as the different ways to measure this type of competence. Having moral judgment is the most popular element of moral competence, but the literature introduces many other elements. The review also shows there is a plethora of ways to measure moral competence, either in standardized tests providing scores or other non-standardized tests. As a (...) second task, the article reflects critically on the general use of the idea of moral competence. Results suggest that this idea functions as what philosopher Bernard Williams would have called a thin ethical concept. Thin concepts are not problematic in and of themselves if they are used as linguistic shorthand, however, there may be shortcomings in the literature on moral competence because the idea is not rooted in more substantive views or theories in a way that is both clear and coherent. (shrink)

Social robotics has a high disruptive potential, for it expands the field of application of intelligent technology to practical contexts of a relational nature. Due to their capacity to “intersubjectively” interact with people, social robots can take over new roles in our daily activities, multiplying the ethical implications of intelligent robotics. In this paper, we offer some preliminary considerations for the ethical reflection on social robotics, so that to clarify how to correctly orient the critical-normative thinking in this arduous task. (...) We defend the ‘being perspective’ and its linked categories of ‘teleology’ and ‘interest’ as the ones from which to articulate the reflection. We argue that attending primarily to the ‘being’, before the ‘doing’, of robots allow us to correctly approach the core phenomenon of all ethical concerns on social robotics, avoiding the reductionisms in the ethical gaze to which the latter may lead. (shrink)