The dominant practice of AI alignment assumes (1) that preferences are an adequate representation of human values, (2) that human rationality can be understood in terms of maximizing the satisfaction of preferences, and (3) that AI systems should be aligned with the preferences of one or more humans to ensure that they behave safely and in accordance with our values. Whether implicitly followed or explicitly endorsed, these commitments constitute what we term a preferentist approach to AI alignment. In this paper, (...) we characterize and challenge the preferentist approach, describing conceptual and technical alternatives that are ripe for further research. We first survey the limits of rational choice theory as a descriptive model, explaining how preferences fail to capture the thick semantic content of human values, and how utility representations neglect the possible incommensurability of those values. We then critique the normativity of expected utility theory (EUT) for humans and AI, drawing upon arguments showing how rational agents need not comply with EUT, while highlighting how EUT is silent on which preferences are normatively acceptable. Finally, we argue that these limitations motivate a reframing of the targets of AI alignment: Instead of alignment with the preferences of a human user, developer, or humanity-writ-large, AI systems should be aligned with normative standards appropriate to their social roles, such as the role of a general-purpose assistant. Furthermore, these standards should be negotiated and agreed upon by all relevant stakeholders. On this alternative conception of alignment, a multiplicity of AI systems will be able to serve diverse ends, aligned with normative standards that promote mutual benefit and limit harm despite our plural and divergent values. (shrink)

Increasingly, our cities are confronted with crises. Fuelled by climate change and a loss of biodiversity, increasing inequalities and fragmentation, challenges range from social unrest and outbursts of violence to heatwaves, torrential rainfall, or epidemics. As crises require rapid interventions that overwhelm human decision-making capacity, AI has been portrayed as a potential avenue to support or even automate decision-making. In this paper, I analyse the specific challenges of AI in urban crisis management as an example and test case for many (...) super wicked decision problems. These super wicked problems are characterised by a coincidence of great complexity and urgency. I will argue that from this combination, specific challenges arise that are only partially covered in the current guidelines and standards around trustworthy or human-centered AI. By following a decision-centric perspective, I argue that to solve urgent crisis problems, the context, capacities, and networks need to be addressed. AI for crisis response needs to follow dedicated design principles that ensure (i) human control in complex social networks, where many humans interact with AI; (ii) principled design that considers core principles of crisis response such as solidarity and humanity; (iii) designing for the most vulnerable. As such this paper is meant to inspire researchers, AI developers and practitioners in the space of AI for (urban) crisis response – and other urgent and complex problems that urban planners are confronted with. (shrink)

Ambitious value learning proposals to solve the AI alignment problem and avoid catastrophic outcomes from a possible future misaligned artificial superintelligence (such as Coherent Extrapolated Volition [CEV]) have focused on ensuring that an artificial superintelligence (ASI) would try to do what humans would want it to do. However, present and future sentient non-humans, such as non-human animals and possible future digital minds could also be affected by the ASI’s behaviour in morally relevant ways. This paper puts forward Sentientist Coherent Extrapolated (...) Volition, an alternative to CEV, that directly takes into account the interests of all sentient beings. This ambitious value learning proposal would significantly reduce the likelihood of risks of astronomical suffering from the ASI’s behaviour, and thus we have very strong pro-tanto moral reasons in favour of implementing it instead of CEV. This fact is crucial in conducting an adequate cost-benefit analysis between different ambitious value learning proposals. (shrink)

Artificially intelligent systems will be used to make increasingly important decisions about us. Many of these decisions will have to be made without universal agreement about the relevant moral facts. For other kinds of disagreement, it is at least usually obvious what kind of solution is called for. What makes moral disagreement especially challenging is that there are three different ways of handling it. _Moral solutions_ apply a moral theory or related principles and largely ignore the details of the disagreement. (...) _Compromise solutions_ apply a method of finding a compromise and taking information about the disagreement as input. _Epistemic solutions_ apply an evidential rule that treats the details of the disagreement as evidence of moral truth. Proposals for all three kinds of solutions can be found in the AI ethics and value alignment literature, but little has been said to justify choosing one over the other. I argue that the choice is best framed in terms of _moral risk_. (shrink)

Value Sensitive Design (VSD) is a design methodology developed by Batya Friedman and Peter Kahn (2003) that brings in moral deliberations in an early stage of a design process. It assumes that neither technology itself is value neutral, nor shifts the value-ladennes to the sole usage of technology. This paper adds to emerging literature onVSD for autonomous weapons systems development and discusses extant literature on values in autonomous systems development in general and in autonomous weapons development in particular. I identify (...) opportunities, such as public debates, and threats, such as the classified nature of the design process, for VSD in autonomous weapons development. This paper contributes to academic debates about the non-neutrality of technology by pointing out that values have been and can be explicitly designed into autonomous systems. It is informative for policy makers and designers who are tasked with developing actual autonomous weapons or policies around such systems, as they learn about an established design methodology that is sensitive to societal concerns and industry needs and that can be applied to autonomous weapons systems. (shrink)

Organisations that design and deploy artificial intelligence (AI) systems increasingly commit themselves to high-level, ethical principles. However, there still exists a gap between principles and practices in AI ethics. One major obstacle organisations face when attempting to operationalise AI Ethics is the lack of a well-defined material scope. Put differently, the question to which systems and processes AI ethics principles ought to apply remains unanswered. Of course, there exists no universally accepted definition of AI, and different systems pose different ethical (...) challenges. Nevertheless, pragmatic problem-solving demands that things should be sorted so that their grouping will promote successful actions for some specific end. In this article, we review and compare previous attempts to classify AI systems for the purpose of implementing AI governance in practice. We find that attempts to classify AI systems proposed in previous literature use one of three mental models: _the Switch_, i.e., a binary approach according to which systems either are or are not considered AI systems depending on their characteristics; _the Ladder_, i.e., a risk-based approach that classifies systems according to the ethical risks they pose; and _the Matrix_, i.e., a multi-dimensional classification of systems that take various aspects into account, such as context, input data, and decision-model. Each of these models for classifying AI systems comes with its own set of strengths and weaknesses. By conceptualising different ways of classifying AI systems into simple mental models, we hope to provide organisations that design, deploy, or regulate AI systems with the vocabulary needed to demarcate the material scope of their AI governance frameworks. (shrink)

This paper further cashes out the notion that particular types of intelligent systems are susceptible to the is-ought problem, which espouses the thesis that no evaluative conclusions may be inferred from factual premises alone. Specifically, it focuses on top-down artificial moral agents, providing ancillary support to the view that these kinds of artifacts are not capable of producing genuine moral judgements. Such is the case given that machines built via the classical programming approach are always composed of two parts, namely: (...) a world model and utility function. In principle, any attempt to bridge the gap between these two would fail, since their reconciliation necessitates for the derivation of evaluative claims from factual premises. (shrink)

Recent work in AI ethics often calls for AI to advance human values and interests. The concept of “AI for people” is one notable example. Though commendable in some respects, this work falls short by excluding the moral significance of nonhumans. This paper calls for a shift in AI ethics to more inclusive paradigms such as “AI for the world” and “AI for the universe”. The paper outlines the case for more inclusive paradigms and presents implications for moral philosophy and (...) computer science work on AI ethics. (shrink)

As the range of potential uses for Artificial Intelligence, in particular machine learning, has increased, so has awareness of the associated ethical issues. This increased awareness has led to the realisation that existing legislation and regulation provides insufficient protection to individuals, groups, society, and the environment from AI harms. In response to this realisation, there has been a proliferation of principle-based ethics codes, guidelines and frameworks. However, it has become increasingly clear that a significant gap exists between the theory of (...) AI ethics principles and the practical design of AI systems. In previous work, we analysed whether it is possible to close this gap between the ‘what’ and the ‘how’ of AI ethics through the use of tools and methods designed to help AI developers, engineers, and designers translate principles into practice. We concluded that this method of closure is currently ineffective as almost all existing translational tools and methods are either too flexible or too strict. This raised the question: if, even with technical guidance, AI ethics is challenging to embed in the process of algorithmic design, is the entire pro-ethical design endeavour rendered futile? And, if no, then how can AI ethics be made useful for AI practitioners? This is the question we seek to address here by exploring why principles and technical translational tools are still needed even if they are limited, and how these limitations can be potentially overcome by providing theoretical grounding of a concept that has been termed ‘Ethics as a Service.’. (shrink)

Alan Turing, one of the fathers of computing, warned that Artificial Intelligence (AI) could one day pose an existential risk to humanity. Today, recent advancements in the field AI have been accompanied by a renewed set of existential warnings. But what exactly constitutes an existential risk? And how exactly does AI pose such a threat? In this chapter we aim to answer these questions. In particular, we will critically explore three commonly cited reasons for thinking that AI poses an existential (...) threat to humanity: the control problem, the possibility of global disruption from an AI race dynamic, and the weaponization of AI. (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 idea of artificial intelligence for social good is gaining traction within information societies in general and the AI community in particular. It has the potential to tackle social problems through the development of AI-based solutions. Yet, to date, there is only limited understanding of what makes AI socially good in theory, what counts as AI4SG in practice, and how to reproduce its initial successes in terms of policies. This article addresses this gap by identifying seven ethical factors that are (...) essential for future AI4SG initiatives. The analysis is supported by 27 case examples of AI4SG projects. Some of these factors are almost entirely novel to AI, while the significance of other factors is heightened by the use of AI. From each of these factors, corresponding best practices are formulated which, subject to context and balance, may serve as preliminary guidelines to ensure that well-designed AI is more likely to serve the social good. (shrink)

Responsible innovation in artificial intelligence calls for public deliberation: well-informed “deep democratic” debate that involves actors from the public, private, and civil society sectors in joint efforts to critically address the goals and means of AI. Adopting such an approach constitutes a challenge, however, due to the opacity of AI and strong knowledge boundaries between experts and citizens. This undermines trust in AI and undercuts key conditions for deliberation. We approach this challenge as a problem of situating the knowledge of (...) actors from the AI industry within a deliberative system. We develop a new framework of responsibilities for AI innovation as well as a deliberative governance approach for enacting these responsibilities. In elucidating this approach, we show how actors from the AI industry can most effectively engage with experts and nonexperts in different social venues to facilitate well-informed judgments on opaque AI systems and thus effectuate their democratic governance. (shrink)

Public consumption of artificial intelligence (AI) technologies has been rarely investigated from the perspective of data surveillance and security. We show that the technology acceptance model, when properly modified with security and surveillance fears about AI, builds an insight on how individuals begin to use, accept, or evaluate AI and its automated decisions. We conducted two studies, and found positive roles of perceived ease of use (PEOU) and perceived usefulness (PU). AI security concern, however, negatively affected PEOU and PU, resulting (...) in less acceptance of AI—(1) use, (2) preference, and (3) participation. AI surveillance concern also had negative effects on the credibility of AI and its recommendations. We integrated extant literature on socio-demographic differences, providing an insight on how AI acceptance is based on one’s rationality regarding (1) technological risks (security/surveillance) and (2) benefits (PEOU/PU) as well as other contextual factors of socio-demographics. (shrink)

As the range of potential uses for Artificial Intelligence (AI), in particular machine learning (ML), has increased, so has awareness of the associated ethical issues. This increased awareness has led to the realisation that existing legislation and regulation provides insufficient protection to individuals, groups, society, and the environment from AI harms. In response to this realisation, there has been a proliferation of principle-based ethics codes, guidelines and frameworks. However, it has become increasingly clear that a significant gap exists between the (...) theory of AI ethics principles and the practical design of AI systems. In previous work , we analysed whether it is possible to close this gap between the ‘what’ and the ‘how’ of AI ethics through the use of tools and methods designed to help AI developers, engineers, and designers translate principles into practice. We concluded that this method of closure is currently ineffective as almost all existing translational tools and methods are either too flexible (and thus vulnerable to ethics washing) or too strict (unresponsive to context). This raised the question: if, even with technical guidance, AI ethics is challenging to embed in the process of algorithmic design, is the entire pro-ethical design endeavour rendered futile? And, if no, then how can AI ethics be made useful for AI practitioners? This is the question we seek to address here by exploring why principles and technical translational tools are still needed even if they are limited, and how these limitations can be potentially overcome by providing theoretical grounding of a concept that has been termed ‘Ethics as a Service’. (shrink)

To be intrinsically valuable means to be valuable for its own sake. Moral philosophy is often ethically anthropocentric, meaning that it locates intrinsic value within humans. This paper rejects ethical anthropocentrism and asks, in what ways might nonhumans be intrinsically valuable? The paper answers this question with a wide-ranging survey of theories of nonhuman intrinsic value. The survey includes both moral subjects and moral objects, and both natural and artificial nonhumans. Literatures from environmental ethics, philosophy of technology, philosophy of art, (...) moral psychology, and related fields are reviewed, and gaps in these literatures are identified. Although the gaps are significant and much work remains to be done, the survey nonetheless demonstrates that those who reject ethical anthropocentrism have considerable resources available to develop their moral views. Given the many very high-stakes issues involving both natural and artificial nonhumans, and the sensitivity of these issues to how nonhumans are intrinsically valued, this is a vital project to pursue. (shrink)

Artificial intelligence (AI) is becoming part of the everyday. During this transition, people’s intention to use AI technologies is still unclear and emotions such as fear are influencing it. In this paper, we focus on autonomous cars to first verify empirically the extent to which people fear AI and then examine the impact that fear has on their intention to use AI-driven vehicles. Our research is based on a systematic survey and it reveals that while individuals are largely afraid of (...) cars that are driven by AI, they are nonetheless willing to adopt this technology as soon as possible. To explain this tension, we extend our analysis beyond just fear and show that people also believe that AI-driven cars will generate many individual, urban and global benefits. Subsequently, we employ our empirical findings as the foundations of a theoretical framework meant to illustrate the main factors that people ponder when they consider the use of AI tech. In addition to offering a comprehensive theoretical framework for the study of AI technology acceptance, this paper provides a nuanced understanding of the tension that exists between the fear and adoption of AI, capturing what exactly people fear and intend to do. (shrink)

As artificial intelligence (AI) continues to proliferate into every area of modern life, there is no doubt that society has to think deeply about the potential impact, whether negative or positive, that it will have. Whilst scholars recognise that AI can usher in a new era of personal, social and economic prosperity, they also warn of the potential for it to be misused towards the detriment of society. Deliberate strategies are therefore required to ensure that AI can be safely integrated (...) into society in a manner that would maximise the good for as many people as possible, whilst minimising the bad. One of the most urgent societal expectations of artificial agents is the need for them to behave in a manner that is morally relevant, i.e. to become artificial moral agents (AMAs). In this article, I will argue that exemplarism, an ethical theory based on virtue ethics, can be employed in the building of computationally rational AMAs with weak machine ethics. I further argue that three features of exemplarism, namely grounding in moral exemplars, meeting community expectations and practical simplicity, are crucial to its uniqueness and suitability for application in building AMAs that fit the ethos of AI4SG. (shrink)

Organisations increasingly use automated decision-making systems (ADMS) to inform decisions that affect humans and their environment. While the use of ADMS can improve the accuracy and efficiency of decision-making processes, it is also coupled with ethical challenges. Unfortunately, the governance mechanisms currently used to oversee human decision-making often fail when applied to ADMS. In previous work, we proposed that ethics-based auditing (EBA)—that is, a structured process by which ADMS are assessed for consistency with relevant principles or norms—can (a) help organisations (...) verify claims about their ADMS and (b) provide decision-subjects with justifications for the outputs produced by ADMS. In this article, we outline the conditions under which EBA procedures can be feasible and effective in practice. First, we argue that EBA is best understood as a ‘soft’ yet ‘formal’ governance mechanism. This implies that the main responsibility of auditors should be to spark ethical deliberation at key intervention points throughout the software development process and ensure that there is sufficient documentation to respond to potential inquiries. Second, we frame AMDS as parts of larger sociotechnical systems to demonstrate that to be feasible and effective, EBA procedures must link to intervention points that span all levels of organisational governance and all phases of the software lifecycle. The main function of EBA should, therefore, be to inform, formalise, assess, and interlink existing governance structures. Finally, we discuss the policy implications of our findings. To support the emergence of feasible and effective EBA procedures, policymakers and regulators could provide standardised reporting formats, facilitate knowledge exchange, provide guidance on how to resolve normative tensions, and create an independent body to oversee EBA of ADMS. (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)

This article is about a specific, but so far neglected peril of AI, which is that AI systems may become existential as well as causing suffering risks for nonhuman animals. The AI value alignment problem has now been acknowledged as critical for AI safety as well as very hard. However, currently it has only been attempted to align the values of AI systems with human values. It is argued here that this ought to be extended to the values of nonhuman (...) animals since it would be speciesism not to do so. The article focuses on the two subproblems—value extraction and value aggregation—discusses challenges for the integration of values of nonhuman animals and explores approaches to how AI systems could address them. (shrink)

This chapter argues that dual-use emerging technologies are distributing unprecedented offensive capabilities to nonstate actors. To counteract this trend, some scholars have proposed that states become a little “less liberal” by implementing large-scale surveillance policies to monitor the actions of citizens. This is problematic, though, because the distribution of offensive capabilities is also undermining states’ capacity to enforce the rule of law. I will suggest that the only plausible escape from this conundrum, at least from our present vantage point, is (...) the creation of a “supersingleton” run by a friendly superintelligence, founded upon a “post-singularity social contract.” In making this argument, the present chapter offers a novel reason for prioritizing the “control problem,” i.e., the problem of ensuring that a greaterthan-human-level AI will positively enhance human well-being. (shrink)

Characterizing human values is a topic deeply interwoven with the sciences, humanities, political philosophy, art, and many other human endeavors. In recent years, a number of thinkers have argued that accelerating trends in computer science, cognitive science, and related disciplines foreshadow the creation of intelligent machines which meet and ultimately surpass the cognitive abilities of human beings, thereby entangling an understanding of human values with future technological development. Contemporary research accomplishments suggest increasingly sophisticated AI systems becoming widespread and responsible for (...) managing many aspects of the modern world, from preemptively planning users’ travel schedules and logistics, to fully autonomous vehicles, to domestic robots assisting in daily living. The extrapolation of these trends has been most forcefully described in the context of a hypothetical “intelligence explosion,” in which the capabilities of an intelligent software agent would rapidly increase due to the presence of feedback loops unavailable to biological organisms. The possibility of superintelligent agents, or simply the widespread deployment of sophisticated, autonomous AI systems, highlights an important theoretical problem: the need to separate the cognitive and rational capacities of an agent from the fundamental goal structure, or value system, which constrains and guides the agent’s actions. The “value alignment problem” is to specify a goal structure for autonomous agents compatible with human values. In this brief article, we suggest that ideas from affective neuroscience and related disciplines aimed at characterizing neurological and behavioral universals in the mammalian kingdom provide important conceptual foundations relevant to describing human values. We argue that the notion of “mammalian value systems” points to a potential avenue for fundamental research in AI safety and AI ethics. (shrink)