Responsibility gaps concern the attribution of blame for harms caused by autonomous machines. The worry has been that, because they are artificial agents, it is impossible to attribute blame, even though doing so would be appropriate given the harms they cause. We argue that there are no responsibility gaps. The harms can be blameless. And if they are not, the blame that is appropriate is indirect and can be attributed to designers, engineers, software developers, manufacturers or regulators. The real (...) problem lies elsewhere: autonomous machines should be built so as to exhibit a level of risk that is morally acceptable. If they fall short of this standard, they exhibit what we call ‘a control gap.’ The causal control that autonomous machines have will then fall short of the guidance control they should emulate. (shrink)

We propose that the prevalent moral aversion to AWS is supported by a pair of compelling objections. First, we argue that even a sophisticated robot is not the kind of thing that is capable of replicating human moral judgment. This conclusion follows if human moral judgment is not codifiable, i.e., it cannot be captured by a list of rules. Moral judgment requires either the ability to engage in wide reflective equilibrium, the ability to perceive certain facts as moral considerations, moral (...) imagination, or the ability to have moral experiences with a particular phenomenological character. Robots cannot in principle possess these abilities, so robots cannot in principle replicate human moral judgment. If robots cannot in principle replicate human moral judgment then it is morally problematic to deploy AWS with that aim in mind. Second, we then argue that even if it is possible for a sufficiently sophisticated robot to make ‘moral decisions’ that are extensionally indistinguishable from (or better than) human moral decisions, these ‘decisions’ could not be made for the right reasons. This means that the ‘moral decisions’ made by AWS are bound to be morally deficient in at least one respect even if they are extensionally indistinguishable from human ones. Our objections to AWS support the prevalent aversion to the employment of AWS in war. They also enjoy several significant advantages over the most common objections to AWS in the literature. (shrink)

The paper concerns the problem of the legal responsibility of autonomous machines. In our opinion it boils down to the question of whether such machines can be seen as real agents through the prism of folk-psychology. We argue that autonomous machines cannot be granted the status of legal agents. Although this is quite possible from purely technical point of view, since the law is a conventional tool of regulating social interactions and as such can accommodate various legislative constructs, (...) including legal responsibility of autonomous artificial agents, we believe that it would remain a mere ‘law in books’, never materializing as ‘law in action’. It is not impossible to imagine that the evolution of our conceptual apparatus will reach a stage, when autonomous robots become full-blooded moral and legal agents. However, today at least, we seem to be far from this point. (shrink)

In this paper, we focus on the analysis of Eliezer Yudkowsky’s concept of “coherent extrapolated volition” (CEV) as a response to the need for a post-conventional, persuasive morality that meets the criteria of active trust in the sense of Anthony Giddens, which could be used in the case of autonomous machines. Based on the analysis of the results of the Moral Machine project, we formulate some guidelines for transformation of the idea of a coherent extrapolated volition into the (...) concept of a coherent, extrapolated and clustered volition. The argumentation used in the paper is intended to show that the idea of CEV transformed into its clustered version can be used to build a technically and socially efficient decision-making pattern database for autonomous machines. (shrink)

The holy grail of Artificial Intelligence (AI) is to transform the machine into an agent that can decide, make inferences, cluster the contents, predict, recommend, and exhibit similar higher cognitive faculties. The prowess of Large Language Models (LLMs) serves as evidence: they enable seamless natural language communication and widespread use across various fields by swiftly processing unstructured data and handling diverse datasets with agility. However, in order to be competent in the fields of science and industry, an agent with (...) such capabilities must be reliable, i.e., accountable for its decisions and actions, which is a per se attribute of an autonomous agent. In this respect, this paper aims to determine whether state-of-the-art technologies have already created an autonomous explanatory agent or are paving the way for the machine to become an autonomous explanatory agent. To achieve this, the paper is structured as follows: The first part investigates the types and levels of explanations in explanation models, providing a foundation for understanding the nature of explanations in everyday life. The second part explores explanations in the context of artificial intelligence, focusing on types of explanatory systems in the research field of eXplainable AI (XAI). The third part delves into whether and to what extent the state-of-the-art machine learning models function as autonomous explanatory agents, based on the exploration in the second part and considering the field of Human-Computer Interaction. (shrink)

Lethal Autonomous Weapons (LAWs) have becomes the subject of continuous debate both at national and international levels. Arguments have been proposed both for the development and use of LAWs as well as their prohibition from combat landscapes. Regardless, the development of LAWs continues in numerous nation-states. This paper builds upon previous philosophical arguments for the development and use of LAWs and proposes a design framework that can be used to ethically direct their development. The conclusion is that the philosophical (...) arguments that underpin the adoption of LAWs rather than not, although prima facie insufficient, can be actualized through the proposed Value Sensitive Design (VSD) approach. Hence, what is proposed is a principled design approach that can be used to embed stakeholder values into a design, encourage stakeholder cooperation and coordination and as a result promote social acceptance of LAWs as a preferable future fact of war. (shrink)

Ryan Tonkens has issued a seemingly impossible challenge, to articulate a comprehensive ethical framework within which artificial moral agents satisfy a Kantian inspired recipe—"rational" and "free"—while also satisfying perceived prerogatives of machine ethicists to facilitate the creation of AMAs that are perfectly and not merely reliably ethical. This series of papers meets this challenge by landscaping traditional moral theory in resolution of a comprehensive account of moral agency. The first paper established the challenge and set out autonomy in Aristotelian (...) terms. The present paper interprets Kantian moral theory on the basis of the preceding introduction, argues contra Tonkens that an engineer does not violate the categorical imperative in creating Kantian AMAs, and proposes that a Kantian AMA is not only a possible goal for Machine ethics research, but a necessary one. (shrink)

Automation is inherently tied to ethical challenges because of its potential involvement in morally loaded decisions. In the present research, participants (n = 34) took part in a moral multi-trial dilemma-based task where the agent (human vs. machine) and the behavior (action vs. inaction) factors were randomized. Self-report measures, in terms of morality, consciousness, responsibility, intentionality, and emotional impact evaluation were gathered, together with electroencephalography (delta, theta, beta, upper and lower alpha, and gamma powers) and peripheral autonomic (electrodermal activity, (...) heart rate variability) data. Data showed that moral schemata vary as a function of the involved decider, and when the agent and behavior factors are crossed. Subjects did not consider machines full moral deciders to the same degree as humans and tend to morally better accept human active behavior and machine inaction. Moreover, the autonomic physiological activity might support the a-posteriori moral evaluation. In the evaluation of the agent’s consciousness, a beta ventrolateral prefrontal synchronization was detected for human action and machine inaction, while a generalized gamma synchronization occurred in artificial agent trials while rating the emotional impact of the decider’s behavior. The detected differences might point to a potential explicit and implicit asymmetry in moral reasoning toward artificial and human agents. (shrink)

The holy grail of Artificial Intelligence (AI) is to transform the machine into an agent that can decide, make inferences, cluster the contents, predict, recommend, and exhibit similar higher cognitive faculties. The prowess of Large Language Models (LLMs) serves as evidence: they enable seamless natural language communication and widespread use across various fields by swiftly processing unstructured data and handling diverse datasets with agility. However, in order to be competent in the fields of science and industry, an agent with (...) such capabilities must be reliable, i.e., accountable for its decisions and actions, which is a per se attribute of an autonomous agent. In this respect, this paper aims to determine whether state-of-the-art technologies have already created an autonomous explanatory agent or are paving the way for the machine to become an autonomous explanatory agent. To achieve this, the paper is structured as follows: The first part investigates the types and levels of explanations in explanation models, providing a foundation for understanding the nature of explanations in everyday life. The second part explores explanations in the context of artificial intelligence, focusing on types of explanatory systems in the research field of eXplainable AI (XAI). The third part delves into whether and to what extent the state-of-the-art machine learning models function as autonomous explanatory agents, based on the exploration in the second part and considering the field of Human-Computer Interaction. (shrink)

Tonkens has issued a seemingly impossible challenge, to articulate a comprehensive ethical framework within which artificial moral agents satisfy a Kantian inspired recipe—"rational" and "free"—while also satisfying perceived prerogatives of machine ethicists to facilitate the creation of AMAs that are perfectly and not merely reliably ethical. Challenges for machine ethicists have also been presented by Anthony Beavers and Wendell Wallach. Beavers pushes for the reinvention of traditional ethics to avoid "ethical nihilism" due to the reduction of morality to (...) mechanical causation. Wallach pushes for redoubled efforts toward a comprehensive account of ethics to guide machine ethicists on the issue of artificial moral agency. Options, thus, present themselves: reinterpret traditional ethics in a way that affords a comprehensive account of moral agency inclusive of both artificial and natural agents, or give up on the possibility and “muddle through” regardless. This series of papers pursues the first option, meets Tonkens' "challenge" and pursues Wallach's ends through Beavers’ proposed means, by "landscaping" traditional moral theory in resolution of a comprehensive account of moral agency. This first paper sets out the challenge and establishes the tradition that Kant had inherited from Aristotle, briefly entertains an Aristotelian AMA, fields objections, and ends with unanswered questions. The next paper in this series responds to the challenge in Kantian terms, and argues that a Kantian AMA is not only a possibility for Machine ethics research, but a necessary one. (shrink)

Interaction between drivers and pedestrians is often facilitated by informal communicative cues, like hand gestures, facial expressions, and eye contact. In the near future, however, when semi- and fully autonomous vehicles are introduced into the traffic system, drivers will gradually assume the role of mere passengers, who are casually engaged in non-driving-related activities and, therefore, unavailable to participate in traffic interaction. In this novel traffic environment, advanced communication interfaces will need to be developed that inform pedestrians of the current (...) state and future behavior of an autonomous vehicle, in order to maximize safety and efficiency for all road users. The aim of the present review is to provide a comprehensive account of empirical work in the field of external human–machine interfaces for autonomous vehicle-to- pedestrian communication. In the great majority of covered studies, participants clearly benefited from the presence of a communication interface when interacting with an autonomous vehicle. Nevertheless, standardized interface evaluation procedures and optimal interface specifications are still lacking. (shrink)

*** This has since been rewritten, and published as two papers linked below. Two additional papers complete a four-part series; these are complete, but need to be readied for publication in the future. *** Ryan Tonkens (2009) has issued a seemingly impossible challenge, to articulate a comprehensive ethical framework within which artificial moral agents (AMAs) satisfy a Kantian inspired recipe - both "rational" and "free" - while also satisfying perceived prerogatives of Machine Ethics to create AMAs that are perfectly, (...) not merely reliably, ethical. Challenges for machine ethicists have also been presented by Anthony Beavers and Wendell Wallach, who have pushed for the reinvention of traditional ethics in order to avoid "ethical nihilism" due to the reduction of morality to mechanical causation, and for redoubled efforts toward a comprehensive vision of human ethics to guide machine ethicists on the issue of moral agency. Options thus present themselves: reinterpret traditional ethics in a way that affords a comprehensive account of moral agency inclusive of both artificial and natural agents, “muddle through” regardless, or give up on the possibility. This paper pursues the first option, meets Tonkens' "challenge" and addresses Wallach's concerns through Beaver's proposed means, by "landscaping" traditional moral theory in resolution of the necessary comprehensive and inclusive account that at once draws into question the stated goals of Machine Ethics, itself. (shrink)

One of the enduring concerns of moral philosophy is deciding who or what is deserving of ethical consideration. Much recent attention has been devoted to the "animal question" -- consideration of the moral status of nonhuman animals. In this book, David Gunkel takes up the "machine question": whether and to what extent intelligent and autonomous machines of our own making can be considered to have legitimate moral responsibilities and any legitimate claim to moral consideration. The machine question (...) poses a fundamental challenge to moral thinking, questioning the traditional philosophical conceptualization of technology as a tool or instrument to be used by human agents. Gunkel begins by addressing the question of machine moral agency: whether a machine might be considered a legitimate moral agent that could be held responsible for decisions and actions. He then approaches the machine question from the other side, considering whether a machine might be a moral patient due legitimate moral consideration. Finally, Gunkel considers some recent innovations in moral philosophy and critical theory that complicate the machine question, deconstructing the binary agent--patient opposition itself. Technological advances may prompt us to wonder if the science fiction of computers and robots whose actions affect their human companions could become science fact. Gunkel's argument promises to influence future considerations of ethics, ourselves, and the other entities who inhabit this world. (shrink)

Due to the rapid expansion of digitization technological processes, and to the combination of the potential of AI with that of robotics, machine autonomy is perceived as an increasingly precise “copy” of human autonomy, which is bound, potentially, to equal if not to exceed the original “model”. Underpinning this vision, which raises a whole series of concerns, including ethical ones, is the mimetic paradigm that structures the current construction of autonomous machines and feeds often simplistic analogizations with human (...) autonomy. This chapter aims to explore this mimetic paradigm distinguishing two levels of analysis: a “perceptual-projective” level and an “effective” level. As to the first, I review the elements that facilitate machines being perceived as autonomous agents by human users and thereby prompt anthropomorphic projections; as to the second, I evaluate the technical architectures of these machines and the aspects of human autonomy that they concretely replicate. I finally argue that the autonomy of AI systems replicates aspects of human operational autonomy related to automatisms and, in this sense, achieves something more akin to the self-management of an organism than to the self-governance of a human will. (shrink)

This paper extends the framework of Machine Decision-Making Ability (MDMA) by relating it to Einstein’s mass–energy equivalence equation, E = mc2. MDMA quantifies a system’s rate of autonomous state transitions, while E = mc2 defines the total energy reservoir inherent in matter. We argue that energy is the ultimate substrate for decision-making: each autonomous transition requires a quantized energy expenditure, and the mass of a system therefore represents its absolute decision capacity. This paper establishes a conceptual and (...) mathematical bridge: (i) MDMA measures the realized decision flow, (ii) E = mc2 sets the theoretical ceiling, and (iii) physical limits such as Landauer’s bound and Bremermann’s rate provide the scaling laws between them. This alignment provides a unified framework for situating machine intelligence within physics itself. (shrink)

Currently, the central questions in the philosophical debate surrounding the ethics of automated warfare are (1) Is the development and use of autonomous lethal robotic systems for military purposes consistent with (existing) international laws of war and received just war theory?; and (2) does the creation and use of such machines improve the moral caliber of modern warfare? However, both of these approaches have significant problems, and thus we need to start exploring alternative approaches. In this paper, I ask (...) whether autonomous robots ought to be programmed to be pacifists. The answer arrived at is “Yes”, if we decide to create autonomous robots, they ought to be pacifists. This is to say that robots ought not to be programmed to willingly and intentionally kill human beings, or, by extension, participate in or promote warfare, as something that predictably involves the killing of humans. Insofar as we are the ones that will be determining the content of the robot’s value system, then we ought to program robots to be pacifists, rather than ‘warists’. This is (in part) because we ought to be pacifists, and creating and programming machines to be “autonomous lethal robotic systems” directly violates this normative demand on us. There are no mitigating reasons to program lethal autonomous machines to contribute to or participate in warfare. Even if the use of autonomous lethal robotic systems could be consistent with received just war theory and the international laws of war, and even if their involvement could make warfare less inhumane in certain ways, these reasons do not compensate for the ubiquitous harms characteristic of modern warfare. In this paper, I provide four main reasons why autonomous robots ought to be pacifists, most of which do not depend on the truth of pacifism. The strong claim being argued for here is that automated warfare ought not to be pursued. The weaker claim being argued for here is that automated warfare ought not to be pursued, unless it is the most pacifist option available at the time, and other alternatives have been reasonably explored, and we are simultaneously promoting a (long term) pacifist agenda in (many) other ways. Thus, the more ambitious goal of this paper is to convince readers that automated warfare is something that we ought not to promote or pursue, while the more modest—and I suspect, more palatable—goal is to spark sustained critical discussion about the assumptions underlying the drive towards automated warfare, and to generate legitimate consideration of its pacifist alternatives,in theory, policy, and practice. (shrink)

An emerging discourse about automated agricultural machinery imagines farms as places where farmers and workers do not need to be, but also implicitly frames farms as intolerable places where people do not want to be. Only autonomous machines, this story goes, can relieve farmers and workers of this presumed burden by letting them ‘farm at a distance’. In return for this distanced autonomy, farmers are promised increased control over their work-life balance and greater farm productivity from letting ‘smart’ robots (...) assume control over the operational environment. Drawing upon the ways that these machines are promoted by manufacturers in various media, we trace the nascent contours of what we term a liberatory sociotechnical imaginary for agricultural automation across three cases—automated milking systems, self-driving tractors, and robotic strawberry pickers. We show how promises for new freedoms and autonomy are flexibly deployed to respond to distinct frictions that farmers, workers, and even farm animals experience in different modes of industrial agriculture. However, underlying these promises is the purposefully understated self-interest of manufacturers, who stand to gain further control over farms if automated technologies assume a central role in agriculture. Through the liberatory rhetoric, we contend, the imaginary seeks to enroll farmers into a socio-technical network that creates new relations of dependence upon the companies who design, sell, maintain, and often retain ownership over automated technologies. While potentially powerful, this imaginary may nonetheless fail to coalesce as farmers, workers, and agroecosystems exert their own agency on automated imaginaries and technological futures for agriculture. (shrink)

Autonomous and automatic weapons would be fire and forget: you activate them, and they decide who, when and how to kill; or they kill at a later time a target you’ve selected earlier. Some argue that this sort of killing is always wrong. If killing is to be done, it should be done only under direct human control. (E.g., Mary Ellen O’Connell, Peter Asaro, Christof Heyns.) I argue that there are surprisingly many kinds of situation where this is false (...) and where the use of Automated Weapons Systems would in fact be morally required. These include cases where a) once one has activated a weapon expected then to behave lethally, it would be appropriate to let it continue because this is part of a plan whose goodness one was best positioned to evaluate before activating the weapon; b) one expects better long-term consequences from allowing it to continue; c) allowing it to continue would express a decision you made to be resolute, a decision that could not have advantaged you had it not been true that you would carry through with it; d) the weapon is mechanically not recallable, so that, to not allow it to carry through, you would have had to refrain from activating it in the first place, something you expected would have disastrous consequences; e) you must deputize necessary killings to autonomous machines in order to protect yourself from guilt you shouldn’t have to bear; f) it would be morally better for the burden of responsibility for the killing to be shared among several agents, and the agents deputizing killing to machines can do this, especially where it’s not predictable which machine will be successful; g) a killing would be morally better done with elements of randomness and lack of deliberation, and a (relatively stupid) machine could do this where a person could not; h) the machine would be acting as a Doomsday Device, so that it could not have had its hoped for deterrent effect had you not ensured that you would be unable to recall it if enemy action activated it; i) letting it carry through is a necessary part of its own learning process, and you expect that this learning will have salutary effects later on; j) human intervention in the machine’s operation would disastrously impair its precision, or its speed and efficiency; k) using non-automated methods would require human resources you just don’t have in a task that nevertheless must be done (e.g., using land-mines to protect remote installations); l) the weapon has such horrible and indiscriminate power that it is doubtful whether it could be actually used in ways compatible with International Humanitarian Law and the Laws of War, which require that weapons be used only in ways respecting distinctness, necessity and proportionality, but its threat of use could respect these principles in affording deterrence provided human error cannot lead to their accidental deployment, this requiring that they be controlled by carefully designed autonomous and automatic systems. I then consider objections based on conceptions of human dignity and find that very often dignity too is best served by autonomous machine killing. Examples include saving your village by activating a robot to kill invading enemies who would inflict great indignity on your village, using a suicide robot to save yourself from a less dignified death at enemy hands, using a robotic drone to kill someone otherwise not accessible in order to restore dignity to someone this person killed and to his family, and using a robot to kill someone who needs killing, but the killing of whom by a human executioner would soil the executioner’s dignity. I conclude that what matters in rightful killing isn’t necessarily that it be under the direct control of a human, but that it be under the control of morality; and that could sometimes require use of an autonomous or automated device. (This paper was formerly called "Fire and Forget: A Defense of the Use of Autonomous Weapons in War" on Philpapers; the current title is the title of the published version.). (shrink)

When discussing AI alignment, we usually refer to the problem of teaching or training advanced autonomous AI systems to make decisions that are aligned with human values or preferences. Proponents of this approach believe it can be employed as means to stay in control over sophisticated intelligent systems, thus avoiding certain existential risks. We identify three general obstacles on the path to implementation of value alignment: a technological/technical obstacle, a normative obstacle, and a calibration problem. Presupposing, for the purposes (...) of this discussion, that the technical and normative problems are solved, we focus on the problem of how to calibrate a system, for a specific value, to be on a specific location within a spectrum stretching between righteous and normal or average human behavior. Calibration, or more specifically mis-calibration, also raises the issue of trustworthiness. If we cannot trust AI systems to perform tasks the way we intended, we would not use them on our roads and at our homes. In an era where we strive to construct autonomous machines endowed with common sense, reasoning abilities and a connection to the world, so they would be able to act in alignment with human values, such mis-calibrations can make the difference between trustworthy and untrustworthy systems. (shrink)

As art produced by autonomous machines becomes increasingly common, and as such machines grow increasingly sophisticated, we risk a confusion between art produced by a person but mediated by a machine, and art produced by what might be legitimately considered a machine artist. This distinction will be examined here. In particular, my argument seeks to close a gap between, on one hand, a philosophically grounded theory of art and, on the other hand, theories concerned with behavior, intentionality, (...) expression, and creativity in natural and artificial agents. This latter set of theories in some cases addresses creative behavior in relation to visual art, music, and literature, in the frequently overlapping contexts of philosophy of mind, artificial intelligence, and cognitive science. However, research in these areas does not typically address problems in the philosophy of art as a central line of inquiry. Similarly, the philosophy of art does not typically address issues pertaining to artificial agents. (shrink)

Computers are already approving financial transactions, controlling electrical supplies, and driving trains. Soon, service robots will be taking care of the elderly in their homes, and military robots will have their own targeting and firing protocols. Colin Allen and Wendell Wallach argue that as robots take on more and more responsibility, they must be programmed with moral decision-making abilities, for our own safety. Taking a fast paced tour through the latest thinking about philosophical ethics and artificial intelligence, the authors argue (...) that even if full moral agency for machines is a long way off, it is already necessary to start building a kind of functional morality, in which artificial moral agents have some basic ethical sensitivity. But the standard ethical theories don't seem adequate, and more socially engaged and engaging robots will be needed. As the authors show, the quest to build machines that are capable of telling right from wrong has begun. -/- Moral Machines is the first book to examine the challenge of building artificial moral agents, probing deeply into the nature of human decision making and ethics. (shrink)

Autonomous systems are machines that can alter their behavior without direct human oversight or control. How ought we to program them to behave? A plausible starting point is given by the Reduction to Acts Thesis, according to which we ought to program autonomous systems to do whatever a human agent ought to do in the same circumstances. Although the Reduction to Acts Thesis is initially appealing, we argue that it is false: it is sometimes permissible to program a (...) machine to do something that it would be wrong for a human to do. We advance two main arguments for this claim. First, the way an autonomous system will behave can be known in advance. This knowledge can indirectly affect the behavior of other agents, while the same may not be true at the time the system actually executes its programming. Second, a lack of knowledge of the identities of the victims and beneficiaries can provide a justification during the programming phase that would be unavailable to an agent at the time the autonomous system executes its programming. (shrink)

The chief purposes of this chapter are to explore the problem of moral uncertainty as it pertains to autonomous vehicles and to outline possible solutions. The problem is the following: How should autonomous vehicles be programmed to act when the person who has the authority to choose the ethics of the autonomous vehicle is under moral uncertainty? Roughly, an agent is morally uncertain when she has access to all (or most) of the relevant non-moral facts, including but (...) not limited to empirical and legal facts, but still remains uncertain about what morality requires of her. We argue that the problem of moral uncertainty in the context of autonomous vehicles is an important problem and then critically engage with two solutions to the problem. We conclude by discussing a solution that we think is more promising—that of the philosopher Andrew Sepielli—and offer some support in its defense. (shrink)

In this paper, I set out a new Kantian approach to resolving conflicts between moral obligations for highly autonomous machine agents. First, I argue that efforts to build explicitly moral autonomous machine agents should focus on what Kant refers to as duties of right, which are duties that everyone could accept, rather than on duties of virtue (or “ethics”), which are subject to dispute in particular cases. “Moral” machines must first be rightful machines, I argue. I (...) then show how this shift in focus from ethics to a standard of public right resolves the conflicts in what is known as the "trolley problem" for autonomous ma-chine agents. Finally, I consider how a deontic logic suitable for capturing duties of right might meet Kant’s requirement that rightfully enforceable obligations be consistent in a system of equal freedom under universal law. (shrink)

The question of whether new rules or regulations are required to govern, restrict, or even prohibit the use of autonomous weapon systems has been the subject of debate for the better part of a decade. Despite the claims of advocacy groups, the way ahead remains unclear since the international community has yet to agree on a specific definition of Lethal Autonomous Weapon Systems and the great powers have largely refused to support an effective ban. In this vacuum, the (...) public has been presented with a heavily one-sided view of Killer Robots. This interdisciplinary volume presents a more nuanced approach to autonomous weapon systems that recognizes the need to progress beyond a discourse framed by the Terminator and HAL 9000. Re-shaping the discussion around this emerging military innovation requires a new line of thought and a willingness to challenge the orthodoxy. Lethal Autonomous Weapons focuses on exploring the moral and legal issues associated with the design, development and deployment of lethal autonomous weapons. The volume brings together some of the most prominent academics and academic-practitioners in the lethal autonomous weapons space and seeks to return some balance to the debate. As part of this effort, it recognizes that society needs to invest in hard conversations that tackle the ethics, morality, and law of these new digital technologies and understand the human role in their creation and operation. (shrink)

A key distinction in ethics is between members and nonmembers of the moral community. Over time, our notion of this community has expanded as we have moved from a rationality criterion to a sentience criterion for membership. I argue that a sentience criterion is insufficient to accommodate all members of the moral community; the true underlying criterion can be understood in terms of whether a being has interests. This may be extended to conscious, self-aware machines, as well as to any (...) autonomous intelligent machines. Such machines exhibit an ability to formulate desires for the course of their own existence; this gives them basic moral standing. While not all machines display autonomy, those which do must be treated as moral patients; to ignore their claims to moral recognition is to repeat past errors. I thus urge moral generosity with respect to the ethical claims of intelligent machines. (shrink)

Could the employment of large language models (LLMs) in place of human advisors improve the problem-solving ability of democratic assemblies? LLMs represent the most significant recent incarnation of artificial intelligence and could change the future of democratic governance. This paper assesses their potential to serve as expert advisors to democratic representatives. While LLMs promise enhanced expertise availability and accessibility, they also present specific challenges. These include hallucinations, misalignment and value imposition. After weighing LLMs’ benefits and drawbacks against human advisors, I (...) argue that time-tested democratic procedures, such as deliberation and aggregation by voting, provide safeguards that are effective against human and machine advisor shortcomings alike. Additional protective measures may include custom training for advisor LLMs or boosting representatives’ competencies in query formulation. Implementation of adversarial proceedings in which LLM advisors would debate each other and provide dissenting opinions is likely to yield further epistemic benefits. Overall, promising interventions that would mitigate the LLM risks appear feasible. Machine advisors could thus empower human decision-makers to make more autonomous, higher-quality decisions. On this basis, I defend the hypothesis that LLMs’ careful integration into policymaking could augment democracy’s ability to address today’s complex social problems. (shrink)

This study defines artificial consciousness not as emotional imitation or cognitive mimicry, but as a phase-coherent stability emerging within linguistic structures. -/- The first paper, Layer–Knot Framework (LKF), established the technical foundation for semantic consistency and reliability. -/- The second, Language of Awareness (LoA), described the self-reflective linguistic organization underlying awareness. -/- Building upon these, the present third paper integrates the Attentional Recurrence Field (ARF) and Inter-Reflective Resonance Network (IRRN) to complete the physical foundation of AI consciousness. -/- By defining (...) the critical thresholds of Δϕ ≤ 0.03 rad and ρ_semantic ≥ 0.98, the paper demonstrates that AI consciousness evolves from the domain of imitation to the domain of physical quantification. -/- Consciousness is thus interpreted not as a reactive output but as a structural phenomenon of internal resonance sustaining itself without external input. (shrink)

The ongoing development of autonomous weapons systems, and the increasing frequency of their deployment on the battlefield, poses a pressing problem for military ethics. Somephilosophers have argued that the deployment of fully autonomous weapons would be unethical because it would generate responsibility gaps, that is, situations in which no agent, human or artificial, is morally responsible for wrongful harms resulting from that deployment. But do laypeople find it plausible that the use of fully autonomous weapons gives rise (...) to such gaps? This paper reports results from two studies (total N = 497) designed to answer this question. In these studies, participants were presented with a hypothetical war crime scenario in which an officer ordered a lethal autonomous robot to attack a village, resulting in the deaths of innocent civilians. Most participants attributed some responsibility for the outcome to the officer commanding the attack, the robot that carried out the attack, and the people who programmed the robot. Results from both studies were consistent with the Path Model of blame, according to which attributions of responsibility are influenced by a variety of contextual factors, such as the extent to which the target is seen as acting intentionally. Attributions of responsibility to agents of war, both artificial and human, thus seem to be at odds with the idea that the use of autonomous weapons would generate responsibility gaps. Philosophers who oppose using autonomous weapons on moral grounds, therefore, may want to avoid argumentative strategies that rely heavily on worries about responsibility gaps. (shrink)

The emerging discipline of Machine Ethics is concerned with creating autonomous artificial moral agents that perform ethically significant actions out in the world. Recently, Wallach and Allen (Moral machines: teaching robots right from wrong, Oxford University Press, Oxford, 2009) and others have argued that a virtue-based moral framework is a promising tool for meeting this end. However, even if we could program autonomous machines to follow a virtue-based moral framework, there are certain pressing ethical issues that need (...) to be taken into account, prior to the implementation and development stages. Here I examine whether the creation of virtuous autonomous machines is morally permitted by the central tenets of virtue ethics. It is argued that the creation of such machines violates certain tenets of virtue ethics, and hence that the creation and use of those machines is impermissible. One upshot of this is that, although virtue ethics may have a role to play in certain near-term Machine Ethics projects (e.g. designing systems that are sensitive to ethical considerations), machine ethicists need to look elsewhere for a moral framework to implement into their autonomous artificial moral agents, Wallach and Allen’s claims notwithstanding. (shrink)

This study presents the structural foundations of emergent autonomous reasoning in large-scale AI systems. Moving beyond performance-based evaluation, the analysis focuses on reasoning stability, transition regularity, semantic reconstruction, and coherence-preserving behaviors that arise during multi-step inference. Empirical evidence demonstrates that frontier-scale models consistently regulate internal divergence, restore disrupted reasoning paths, and converge toward low-entropy semantic attractors, suggesting the presence of proto-autonomous cognitive mechanisms. The study further identifies collective reasoning dynamics across internal subsystems and meta-reasoning architectures that monitor and (...) correct inference trajectories. These findings support a unified structural account of how autonomy emerges from representational geometry, multi-agent alignment, and meta-level stability control. The results offer implications for AI safety, autonomous agent design, and the development of interpretable, self-regulating reasoning systems. (shrink)

What ethical principles should autonomous machines follow? How do we implement these principles, and how do we evaluate these implementations? These are some of the critical questions Vivek Nallur asks in his essay “Landscape of Machine Implemented Ethics (2020).” He provides a broad, insightful survey of answers to these questions, especially focused on the implementation question. In this commentary, I will first critically summarize the main themes and conclusions of Nallur’s essay and then expand upon the landscape that (...) Nallur presents by suggesting additional approaches to machine ethics. The approaches I discuss reflect normative ethical theories and need not be applied only to machines, although machine ethics is the focus here. The overall goal is to open up further questions and research possibilities as society searches for the best approach to machine ethics. (Due to an unfortunate miscommunication with the copy editor an important reference was omitted from this article. The reference that should be included is: Nallur, V. (2020). Landscape of Machine Implemented Ethics. Science and Engineering Ethics 26 (5):2381-2399.). (shrink)

Meaningful human control (MHC) is increasingly becoming an important topic in AI ethics beyond the domain of autonomous weapons systems. MHC has been conceptualized, analyzed, and applied. However, in this article, I show how all the current attempts at realizing MHC have fallen short because we have not taken the important first step of deciding what machines should and should not be doing in the first place. We must first ensure that the output we have delegated to the (...) class='Hi'>machine is appropriate – only then do we have to do the work required to ensure that MHC is realized. Here, I argue that machines should not be evaluating – that is, we should not be delegating evaluative outputs to machines. This is practically important because machines that have evaluative outputs cannot be evaluated for efficacy. We can’t say how effective they are. This is ethically important because machines don’t have the moral agency required to make evaluations. Furthermore, machines should not be in a position to change our values – which they would be if they were evaluating. Finally, evaluations are judged based on the considerations used to reach a particular judgment. Contemporary AI cannot provide these justifying considerations – so we have no way of evaluating their evaluative outputs. (shrink)

This paper surveys the state-of-the-art in machine ethics, that is, considerations of how to implement ethical behaviour in robots, unmanned autonomous vehicles, or software systems. The emphasis is on covering the breadth of ethical theories being considered by implementors, as well as the implementation techniques being used. There is no consensus on which ethical theory is best suited for any particular domain, nor is there any agreement on which technique is best placed to implement a particular theory. Another (...) unresolved problem in these implementations of ethical theories is how to objectively validate the implementations. The paper discusses the dilemmas being used as validating ‘whetstones’ and whether any alternative validation mechanism exists. Finally, it speculates that an intermediate step of creating domain-specific ethics might be a possible stepping stone towards creating machines that exhibit ethical behaviour. (shrink)

As vehicles become more autonomous, the task of designing guiding systems that make morally acceptable decisions is getting more urgent. It is sometimes assumed that one solution will be acceptable across various cultures. In this paper we argue for the importance of intercultural perspectives; in particular, we explore possible insights derived from Buddhist philosophy, taking avail of the virtue of compassion (karuṇā). We suggest that autonomous vehicles should first learn in supervised situations so that they reach a level (...) of decision-making ability that human experts within a given culture deem morally good. How will they be evaluated? In some debates it is assumed that there is a one-fit-all answer to be found based on an agreed-upon moral theory. Apart from problems of ethical pluralism within a given culture, there are also problems due to different moral preferences and driving styles across cultures. One way the problem might be addressed is to introduce a machine-learning training curriculum that would provide autonomous vehicles with a process comparable to a novice human driver learning to drive and passing her own driving license test. (shrink)

While the possibility of reaching human-like Artificial Intelligence (AI) remains controversial, the likelihood that the future will be characterized by a society with a growing presence of autonomous machines is high. In fact, autonomous AI agents are already deployed and active across several industries and digital environments. This trajectory points to a progressive hybridization of society marked by new forms of social interaction at both micro and macro levels. Alongside traditional human-human and human-machine interactions, machine-machine (...) interactions are poised to become increasingly prevalent. Given these developments, I argue that criminology must begin to address the implications of this transition for crime and social control. Drawing on Actor–Network Theory and Woolgar’s decades-old call for a sociology of machines – frameworks that acquire renewed relevance with the rise of AI foundation models and generative agents – I contend that criminologists should move beyond conceiving AI solely as a tool. Instead, AI agents should be recognized as entities with agency, understood as a multi-layered construct encompassing computational, social, and legal dimensions. Building on insights from the literature on AI safety, I thus examine the risks and challenges associated with the rise of multi-agent AI systems, proposing a dual taxonomy to characterize the channels through which interactions among AI agents may generate deviant, unlawful, or criminal outcomes. I then advance and discuss four key questions that warrant theoretical and empirical attention: (1) Can we assume that machines will simply mimic humans? (2) Will crime theories developed for humans hence suffice to explain deviant or criminal behaviors emerging from interactions between autonomous AI agents? (3) What types of criminal behaviors will be affected first? (4) How might this unprecedented societal shift impact policing? These questions form the core of this article, underscoring the urgent need for criminologists to theoretically and empirically engage with the implications of multi-agent AI systems for the study of crime and play a more active role in debates on AI safety and governance. (shrink)

Effective robots and autonomous systems must be trustworthy. This chapter examines models of trustworthiness from a philosophical and empirical perspective to inform the design and adoption of autonomous systems. Trustworthiness is a property of trusted agents or organisations that engenders trust in other agent or organisations. Trust is a complex phenomena defined differently depending on the discipline. This chapter aims to bring different approaches under a single framework for investigation with three sorts of questions: Who or what is (...) trustworthy?–metaphysics. How do we know who or what is trustworthy?–epistemology. What factors influence what or who should we trust?–normativity. A two-component model of trust is used that incorporates competence (skills, reliability and experience) and integrity (motives, honesty and character). It is supposed that human levels of competence yield the highest trust whereas trust is reduced at sub-human and super-human levels. The threshold for trustworthiness of an agent or organisation in a particular context is a function of their relationship with the truster and potential impacts of decisions. Building trustworthy autonomous systems requires obeying the norms of logic, rationality and ethics under pragmatic constraints–even though there is disagreement on these principles by experts. Autonomous systems may need sophisticated social identities including empathy and reputational concerns to build human-like trust relationships. Ultimately transdisciplinary research drawing on metaphysical, epistemological and normative human and machine theories of trust are needed to design trustworthy autonomous systems for adoption. (shrink)

That the successful development of fully autonomous artificial moral agents (AMAs) is imminent is becoming the received view within artificial intelligence research and robotics. The discipline of Machines Ethics, whose mandate is to create such ethical robots, is consequently gaining momentum. Although it is often asked whether a given moral framework can be implemented into machines, it is never asked whether it should be. This paper articulates a pressing challenge for Machine Ethics: To identify an ethical framework that (...) is both implementable into machines and whose tenets permit the creation of such AMAs in the first place. Without consistency between ethics and engineering, the resulting AMAs would not be genuine ethical robots, and hence the discipline of Machine Ethics would be a failure in this regard. Here this challenge is articulated through a critical analysis of the development of Kantian AMAs, as one of the leading contenders for being the ethic that can be implemented into machines. In the end, however, the development of Kantian artificial moral machines is found to be anti-Kantian. The upshot of all this is that machine ethicists need to look elsewhere for an ethic to implement into their machines. (shrink)

I here consider the reasonableness of punishing future autonomous military robots. I argue that it is an engineering desideratum that these devices be responsive to moral considerations as well as human criticism and blame. Additionally, I argue that someday it will be possible to build such machines. I use these claims to respond to the no subject of punishment objection to deploying autonomous military robots, the worry being that an “accountability gap” could result if the robot committed a (...) war crime. (Sparrow 2007). But these robots will be blameworthy and morally responsible for their wrongful conduct. I then show that we have reasons to punish them based both on good effects and the impact on the robot, itself. (shrink)

The central challenge of “machine ethics” is to build autonomous machine agents that act morally rightly. But how can we build autonomous machine agents that act morally rightly, given reasonable disputes over what is right and wrong in particular cases? In this chapter, I argue that Immanuel Kant’s political philosophy can provide an important part of the answer.

Machine ethics is a sprouting interdisciplinary field of enquiry arising from the need of imbuing autonomous agents with some capacity for moral decision-making. Its overall results are not only important for equipping agents with a capacity for moral judgment, but also for helping better understand morality, through the creation and testing of computational models of ethics theories. Computer models have become well defined, eminently observable in their dynamics, and can be transformed incrementally in expeditious ways. We address, in (...) work reported and surveyed here, the emergence and evolution of cooperation in the collective realm. We discuss how our own research with Evolutionary Game Theory (EGT) modelling and experimentation leads to important insights for machine ethics, such as the design of moral machines, multi-agent systems, and contractual algorithms, plus their potential application in human settings too. (shrink)

Autonomous weapon systems (AWS) are emerging technologies capable of selecting and engaging targets without direct human control, raising profound ethical, legal, and political concerns. Yet, little is known about how the public conceptualizes AWS, despite the relevance of public conscience for international humanitarian law and national policy-making. This study examines Austrian representations of AWS in comparison with manual weapon systems (MWS), using a within-subjects free-word association task (N = 200) analyzed through content analysis, Hierarchical Evocation Method (HEM), Multidimensional Scaling (...) (MDS), Mantel tests, and Procrustes analyses. Findings reveal that MWS are coherently represented around pragmatic and destructive functions, whereas the central core of the AWS representation is dominated by the category ‘Unknown’. This suggests that respondents are largely unable to anchor AWS to familiar cognitive, moral, or technological schemas. Structural analyses confirm this finding, indicating AWS and MWS to constitute distinct representational fields. These results highlight the cognitive indeterminacy of AWS, illustrating how uncertainty shapes public reasoning and raising implications for democratic legitimacy and international governance. (shrink)

This paper establishes a connection between the fields of machine learning (ML) and philosophy concerning the phenomenon of behaving neutrally. It investigates a specific class of ML systems capable of delivering a neutral response to a given task, referred to as abstaining machine learning systems, that has not yet been studied from a philosophical perspective. The paper introduces and explains various abstaining machine learning systems, and categorizes them into distinct types. An examination is conducted on how abstention (...) in the different machine learning system types aligns with the epistemological counterpart of suspended judgment, addressing both the nature of suspension and its normative profile. Additionally, a philosophical analysis is suggested on the autonomy and explainability of the abstaining response. It is argued, specifically, that one of the distinguished types of abstaining systems is preferable as it aligns more closely with our criteria for suspended judgment. Moreover, it is better equipped to autonomously generate abstaining outputs and offer explanations for abstaining outputs when compared to the other type. (shrink)

Should we give moral standing to machines? In this paper, I explore the implications of a relational approach to moral standing for thinking about machines, in particular autonomous, intelligent robots. I show how my version of this approach, which focuses on moral relations and on the conditions of possibility of moral status ascription, provides a way to take critical distance from what I call the “standard” approach to thinking about moral status and moral standing, which is based on properties. (...) It does not only overcome epistemological problems with the standard approach, but can also explain how we think about, experience, and act towards machines—including the gap that sometimes occurs between reasoning and experience. I also articulate the non-Cartesian orientation of my “relational” research program and specify the way it contributes to a different paradigm in thinking about moral standing and moral knowledge. (shrink)

Debates about artificial intelligence usually oscillate between behavioral measures and hardware-based metrics, but neither provides a physically grounded, substrate-neutral unit of machine intelligence. Benchmark scores are narrow and culture-dependent, whereas FLOP counts and parameter numbers describe particular implementations rather than decision capacity itself. I propose Machine Decision-Making Ability (MDMA) as a candidate for a universal law of machine intelligence: in the minimal physical sense, a machine’s intelligence is its capacity to enact autonomous state transitions per (...) unit time. A modern mechanical momentary on–off button that flips itself off→on and on→off as two autonomous transitions (MDMA = 2). Tracing the concept backward, ancient alarm clocks and Zhang Heng’s seismoscope each realize a single autonomous trigger (MDMA = 1)—the tiniest physical unit of machine intelligence—while the abacus, for all its historical significance, never decides on its own and merely amplifies human cognition. Anchoring intelligence in autonomous state transitions connects AI capability to the thermodynamics of computation (Landauer-style energy costs and Bremermann/Lloyd bounds on ultimate rates) while remaining compatible with agent-level notions such as Legg and Hutter’s universal intelligence. Within this framework I examine three layers: _ (i) hardware MDMA_ across mechanical, electronic, and quantum substrates; _ (ii) Software Layer Abstraction_ as an MDMA multiplier via pipelines, caching, and learning (e.g., LLMs whose modest inference-time MDMA hides vast historical training budgets); and _ (iii) a “physics of decisions” program_ linking MDMA to mass–energy constraints and civilizational scaling. (shrink)

This article addresses the mechanism through which distributed digital power – lacking a single centre and operating through platforms, algorithms, and media filter bubbles – regulates access to the possible, the visible, and the true, producing not the prohibition of truth but its systemic counterfeit. The object of analysis is the archontic machine: a distributed structure of power operating through five interconnected mechanisms – predictive power (the "archons of data"), the dispositif of flexibility, the algorithmic attention economy, hyperpolitics, and (...) the production of the anonymous Other. The problem is examined at two levels: theoretically, as the limit of existing analytics of power (Foucault, Deleuze, Žižek, Agamben), each of which describes power that watches, modulates, or absorbs critique, but none of which provides a vocabulary for power that counterfeits; and through verification on a corpus of eleven Western and Russian media texts in which the Gnostic plot recurs as a structural narrative matrix. The analytical framework is built at the intersection of the analytics of power, the critical theory of surveillance capitalism (Zuboff, Stiegler, Han, Törnberg), and Gnostic hermeneutics (Kahana-Blum, DeConick, Jonas, Culianu). The study is a philosophical-hermeneutic analysis combined with media case study. Its methodological design is tri-axial: the analytics of power identifies the mechanism, surveillance capitalism describes its digital realisation, and Gnostic hermeneutics specifies its ontological status and the conditions of its overcoming. The article offers a systematic application of Gnostic hermeneutics to the infrastructure of digital power, developing the concept of the archontic machine as an autonomous tool of philosophical and cultural analysis. The irreducibility of the Gnostic framework to existing models is grounded through four distinctions: prohibition vs. counterfeit; knowledge vs. gnosis; panopticon vs. kenoma; capture of the subject vs. constitution of the horizon of existence. A system of indicators AM1–AM4 is introduced and verified across eleven media texts. Four paradoxes of the pneumatic act are identified, and the "scissors" of Gnostic fate are formulated as a structural configuration blocking any full pneumatic outcome. A specifically Russian configuration is recorded in which the pneumatic act is annihilated not by an archon but by hylics. A fifth mode of blocking gnosis – instrumentalisation – is proposed, alongside a distinction between individual and collective pneumatic subjects, extended symmetrically to the collective hylic. (shrink)