Deep learning is currently the most prominent and widely successful method in artificial intelligence. Despite having played an active role in earlier artificial intelligence and neural network research, philosophers have been largely silent on this technology so far. This is remarkable, given that deep learning neural networks have blown past predicted upper limits on artificial intelligence performance—recognizing complex objects in natural photographs and defeating world champions in strategy games as complex as Go and chess—yet there remains no universally accepted explanation (...) as to why they work so well. This article provides an introduction to these networks as well as an opinionated guidebook on the philosophical significance of their structure and achievements. It argues that deep learning neural networks differ importantly in their structure and mathematical properties from the shallower neural networks that were the subject of so much philosophical reflection in the 1980s and 1990s. The article then explores several different explanations for their success and ends by proposing three areas of inquiry that would benefit from future engagement by philosophers of mind and science. (shrink)

Social isolation and loneliness are ongoing threats to health made worse by the coronavirus disease 2019 pandemic. During the pandemic, half the globe's population have been placed under strict physical distancing orders and many long-term care facilities serving older adults went into lockdown mode, restricting access to all visitors, including family members. Before the pandemic emerged, a 2020 National Academy of Sciences, Engineering and Medicine report warned of the underappreciated adverse effects of social isolation and loneliness on health, especially among (...) older populations. Social isolation and loneliness predict all-cause mortality at rates that rival clinical risk factors, such as obesity and smoking; they are associated with greater incidence of psychological, cognitive, and physical morbidities. This paper sets forth a proposal to design robots to function as companions and friends for socially isolated and lonely older people during pandemic emergencies and in aging societies more generally. “The proposal” section presents and defends the proposal. The “Replies to objections” section answers objections based on coercive design, replacement of humans with robots, privacy incursions, and counterfeit companionship. The “Conclusion” section submits that sociable robots offer a promising avenue for addressing social isolation and loneliness during pandemics and hold promise for aging societies more broadly. (shrink)

With robots increasingly integrated into various areas of life, the question of relationships with them is gaining prominence. Are friendship and partnership with robots possible? While there is already extensive research on relationships with robots, this article critically examines whether the relationship with non-human entities is sufficiently explored on a deeper level, especially in terms of ethical concepts such as autonomy, agency, and responsibility. In robot ethics, ethical concepts and considerations often presuppose properties such as consciousness, sentience, and intelligence, which (...) are exclusively aligned with humans. I will challenge the understanding of these properties by anchoring them in contexts, bodies, and actions. This approach allows to consider the specific ways of being of various human and non-human entities and to identify these properties in non-human entities as well. My “eco-relational” approach posits that it is crucial to consider the relationality with non-human entities such as animals and technology in central ethical concepts from the beginning. This approach reflects the “eco”, the entire house in which we live including animals and robots. To support this, I examine two dominant approaches in robot ethics within the contemporary Western tradition: the “properties approach” and modestly relational approaches. I will develop an eco-relational approach as an alternative. Employing a phenomenological method, I will demonstrate through various examples that our properties and actions are inherently connected with non-human entities. I will show that robots play a central role in our properties and actions, leading to concepts such as hybrid actions and non-human agency. It becomes clear that technology and our relationships with it disrupt traditional ethical concepts. (shrink)

This addendum expands upon the arguments made in the author’s 2020 essay, “Legal Personhood for Artificial Intelligence: Citizenship as the Exception to the Rule”, in an effort to display the significance human augmentation technologies will have on (feasibly) inadvertently providing legal protections to artificial intelligence systems (AIS)—a topic only briefly addressed in that work. It will also further discuss the impacts popular media have on imprinting notions of computerised behaviour and its subsequent consequences on the attribution of legal protections to (...) AIS and on speculative technological advancement that would aid the sophistication of AIS. (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)

Online exam supervision technologies have recently generated significant controversy and concern. Their use is now booming due to growing demand for online courses and for off-campus assessment options amid COVID-19 lockdowns. Online proctoring technologies purport to effectively oversee students sitting online exams by using artificial intelligence systems supplemented by human invigilators. Such technologies have alarmed some students who see them as a “Big Brother-like” threat to liberty and privacy, and as potentially unfair and discriminatory. However, some universities and educators defend (...) their judicious use. Critical ethical appraisal of online proctoring technologies is overdue. This essay provides one of the first sustained moral philosophical analyses of these technologies, focusing on ethical notions of academic integrity, fairness, non-maleficence, transparency, privacy, autonomy, liberty, and trust. Most of these concepts are prominent in the new field of AI ethics, and all are relevant to education. The essay discusses these ethical issues. It also offers suggestions for educational institutions and educators interested in the technologies about the kinds of inquiries they need to make and the governance and review processes they might need to adopt to justify and remain accountable for using online proctoring technologies. The rapid and contentious rise of proctoring software provides a fruitful ethical case study of how AI is infiltrating all areas of life. The social impacts and moral consequences of this digital technology warrant ongoing scrutiny and study. (shrink)

In this paper, I make a novel case for an expansive approach to engineering ethics education, one that regards micro-ethics and macro-ethics as essentially complementary. Although others have voiced support for including macro-ethical reflection within engineering ethics education, I advance a stronger claim, arguing that isolating engineering ethics from macro-level issues risks rendering even micro-ethical inquiry morally meaningless. I divide my proposal into four parts. First, I clarify the distinction between micro-ethics and macro-ethics as I am construing it, defending my (...) characterization against a potential worry. Second, I consider but reject some arguments for a restrictive approach, one that excludes macro-ethical reflection from engineering ethics education. Third, I offer my central argument for an expansive approach. Finally, I suggest that macro-ethics education can learn something valuable from micro-ethics pedagogy. On my proposal, students consider both micro- and macro-ethical problems from the deliberative perspective, situating micro-ethical problems within a broader social framework but also situating macro-ethical problems within an engaged, practical framework. By emphasizing the value of the deliberative perspective, my proposal contributes to a growing call to broaden the scope of engineering ethics education while maintaining its practical relevance. (shrink)

This study aims to identify artificial intelligence (AI) technologies that are applied in climate-smart agricultural practices and address ethical concerns of deploying those technologies from legal perspectives. As climate-smart agricultural AI, the study considers those AI-based technologies that are used for precision agriculture, monitoring peat lands, deforestation tracking, and improved forest management. The study utilized a systematic literature review approach to identify and analyze AI technologies employed in climate-smart agriculture and associated ethical and legal concerns. The study findings indicate several (...) ethical concerns for deploying AI in climate-smart agricultural practices pertaining to data inaccuracy, other technical errors based on wrong recommendations or wrongful acts, data ownership and intellectual property issues, and economic issues resulting in digital division and privacy and security related issues. In this study, the ethical concerns were further examined based on criminal law, tort law, privacy and data protection law, and intellectual property law. In this regard, the study finds that the current tort law pattern is more suitable than the criminal law pattern to address some major ethical concerns, such as data inaccuracy and other technical errors based on wrong recommendations or wrongful acts. Finally, the study recommends that at the global level, all countries need to fill up the current gap of international law on climate-smart agriculture through agreeing on a standard set of legal provisions and enhancing collaboration in innovation and deployment of climate-smart agricultural AI. It further recommends that at the local level, countries need to adopt suitable regulations addressing multi-stakeholders’ interests associated with the deployment of climate-smart agricultural AIs. (shrink)

Ethics settings allow for morally significant decisions made by humans to be programmed into autonomous machines, such as autonomous vehicles or autonomous weapons. Customizable ethics settings are a type of ethics setting in which the users of autonomous machines make such decisions. Here two arguments are provided in defence of customizable ethics settings. Firstly, by approaching ethics settings in the context of failure management, it is argued that customizable ethics settings are instrumentally and inherently valuable for building resilience into the (...) larger socio-technical systems in which autonomous machines operate. Secondly, after defining the preliminary condition of responsibility attribution and demonstrating how ethics settings enable humans to exert control over the outcomes of morally significant incidents, it is shown that ethics settings narrow the responsibility gap. (shrink)

A encyclopedia entry on the moral impacts that happen when information technologies are used to record, communicate and organize information. including the moral challenges of information technology, specific moral and cultural challenges such as online games, virtual worlds, malware, the technology transparency paradox, ethical issues in AI and robotics, and the acceleration of change in technologies. It concludes with a look at information technology as a model for moral change, moral systems and moral agents.

Robots and other smart machines are increasingly interwoven into the social fabric of our society, with the area and scope of their application continuing to expand. As we become accustomed to interacting through and with robots, we also begin to supplement or replace existing human–human interactions with human–machine interactions. This article aims to discuss the impacts of the shift from human–human interactions to human–machine interactions in one facet of our self-constitution, i.e., morality. More specifically, it sets out to explore whether (...) and how the shift to human–machine interactions can affect our moral cultivation. I shall structure the article around what Shannon Vallor calls technology-driven moral deskilling, i.e., the phenomenon of technology negatively affecting individual moral cultivation, and shall also attempt to offer a Confucian response to the problem. I first elaborate in detail Vallor’s idea of technology-driven moral deskilling. Next, I discuss three paradigms of virtue acquisition identified by Nancy E. Snow, i.e., the “folk” paradigm, the skill-and-expertise paradigm, and the Confucian paradigm, and show how the Confucian paradigm can help us to respond to technology-driven moral deskilling. Finally, I introduce the idea of Confucian rituals (li) and argue for the ritualizing of machines as an answer to technology-driven moral deskilling. (shrink)

Expanding the debate about empathy with human beings, animals, or fictional characters to include human-robot relationships, this paper proposes two different perspectives from which to assess the scope and limits of empathy with robots: the first is epistemological, while the second is normative. The epistemological approach helps us to clarify whether we can empathize with artificial intelligence or, more precisely, with social robots. The main puzzle here concerns, among other things, exactly what it is that we empathize with if robots (...) do not have emotions or beliefs, since they do not have a consciousness in an elaborate sense. However, by comparing robots with fictional characters, the paper shows that we can still empathize with robots and that many of the existing accounts of empathy and mindreading are compatible with such a view. By so doing, the paper focuses on the significance of perspective-taking and claims that we also ascribe to robots something like a perspectival experience. The normative approach examines the moral impact of empathizing with robots. In this regard, the paper critically discusses three possible responses: strategic, anti-barbarizational, and pragmatist. The latter position is defended by stressing that we are increasingly compelled to interact with robots in a shared world and that to take robots into our moral consideration should be seen as an integral part of our self- and other-understanding. (shrink)

Over the past two decades, ethical challenges related to robotics technologies have gained increasing interest among different research and non-academic communities, in particular through the field of roboethics. While the reasons to address roboethics are clear, why not to engage with ethics needs to be better understood. This paper focuses on a limited or lacking engagement with ethics that takes place within some parts of the robotics community and its implications for the conceptualisation of the human being. The underlying assumption (...) is that the term ‘ethical’ essentially means ‘human’. Thus, this paper discusses a working hypothesis according to which by avoiding to engage with roboethics, roboticists contribute to the tacit dehumanisation process emerging in and outside of robotics. An alternative approach includes ‘lived ethics’ which involves not only incorporating formal ethical approaches into the roboticists’ work but also ‘being’ ethical and actually engaging with ethical reflection and practice. (shrink)

With by far the lowest population density in the United States, myriad challenges attach to healthcare delivery in Alaska. In the “Size, Population, and Accessibility” section, we characterize this geographic context, including how it is exacerbated by lack of infrastructure. In the “Distributing Healthcare” section, we turn to healthcare economics and staffing, showing how these bear on delivery—and are exacerbated by geography. In the “Health Care in Rural Alaska” section, we turn to rural care, exploring in more depth what healthcare (...) delivery looks like outside of Alaska’s major cities. This discussion continues in the “Alaska’s Native Villages” section, which specifically analyzes healthcare in Alaska’s indigenous villages, some of the smallest and most isolated communities in the United States. Though many of the ways we could improve Alaskan health care for Alaskan residents are limited by its unique features, the “Justice and Healthcare Delivery” and “Technology and Telemedicine” sections consider ways in which certain policies and technology—including telemedicine—could mitigate the challenges developed in previous sections. (shrink)

Digital transformation and “artificial intelligence (AI)”—which can more adequately be called “data-based systems (DS)”—comprise ethical opportunities and risks. Therefore, it is necessary to identify precisely ethical opportunities and risks in order to be able to benefit sustainably from the opportunities and to master the risks. The UN General Assembly has recently adopted a resolution aiming for ‘safe, secure and trustworthy artificial intelligence systems’. It is now urgent to implement and build on the UN General Assembly Resolution. Allowing humans and the (...) planet to flourish sustainably in peace and guaranteeing globally that human dignity is respected not only offline but also online, in the digital sphere, and in the domain of DS, requires two policy measures: (1) human rights-based data-based systems (HRBDS): HRBDS means that human rights serve as the basis of digital transformation and DS. (2) International Data-Based Systems Agency (IDA): IDA should be established at the UN as a platform for cooperation in the field of digital transformation and DS, fostering human rights, security, and peaceful uses of DS, as well as a global supervisory institution and regulatory authority in digital transformation and DS. The establishment of IDA is realistic because humanity has already shown in its past that we are able to not always “blindly” pursue the technically possible but also to limit ourselves to what is technically feasible when humanity and the planet are at stake. For instance, humans researched the field of nuclear technology, developed the atomic bomb, and detonated it several times. Nonetheless, the same humans limited research and development in the field of nuclear technology to prevent even worse consequences by establishing the International Atomic Energy Agency (IAEA) at the UN. (shrink)

Danaher (2016) has argued that increasing robotization can lead to retribution gaps: Situations in which the normative fact that nobody can be justly held responsible for a harmful outcome stands in conflict with our retributivist moral dispositions. In this paper, we report a cross-cultural empirical study based on Sparrow’s (2007) famous example of an autonomous weapon system committing a war crime, which was conducted with participants from the US, Japan and Germany. We find that (1) people manifest a considerable willingness (...) to hold autonomous systems morally responsible, (2) partially exculpate human agents when interacting with such systems, and that more generally (3) the possibility of normative responsibility gaps is indeed at odds with people’s pronounced retributivist inclinations. We discuss what these results mean for potential implications of the retribution gap and other positions in the responsibility gap literature. (shrink)