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)

With recent advancements in systems engineering and artificial intelligence, autonomous agents are increasingly being called upon to execute tasks that have normative relevance. These are tasks that directly—and potentially adversely—affect human well-being and demand of the agent a degree of normative-sensitivity and -compliance. Such norms and normative principles are typically of a social, legal, ethical, empathetic, or cultural (‘SLEEC’) nature. Whereas norms of this type are often framed in the abstract, or as high-level principles, addressing normative concerns in concrete (...) applications of autonomous agents requires the refinement of normative principles into explicitly formulated practical rules. This paper develops a process for deriving specification rules from a set of high-level norms, thereby bridging the gap between normative principles and operational practice. This enables autonomous agents to select and execute the most normatively favourable action in the intended context premised on a range of underlying relevant normative principles. In the translation and reduction of normative principles to SLEEC rules, we present an iterative process that uncovers normative principles, addresses SLEEC concerns, identifies and resolves SLEEC conflicts, and generates both preliminary and complex normatively-relevant rules, thereby guiding the development of autonomous agents and better positioning them as normatively SLEEC-sensitive or SLEEC-compliant. (shrink)

With recent advancements in systems engineering and artificial intelligence, autonomous agents are increasingly being called upon to execute tasks that have normative relevance. These are tasks that directly—and potentially adversely—affect human well-being and demand of the agent a degree of normative-sensitivity and -compliance. Such norms and normative principles are typically of a social, legal, ethical, empathetic, or cultural nature. Whereas norms of this type are often framed in the abstract, or as high-level principles, addressing normative concerns in concrete applications (...) of autonomous agents requires the refinement of normative principles into explicitly formulated practical rules. This paper develops a process for deriving specification rules from a set of high-level norms, thereby bridging the gap between normative principles and operational practice. This enables autonomous agents to select and execute the most normatively favourable action in the intended context premised on a range of underlying relevant normative principles. In the translation and reduction of normative principles to SLEEC rules, we present an iterative process that uncovers normative principles, addresses SLEEC concerns, identifies and resolves SLEEC conflicts, and generates both preliminary and complex normatively-relevant rules, thereby guiding the development of autonomous agents and better positioning them as normatively SLEEC-sensitive or SLEEC-compliant. (shrink)

인공지능이 자율성을 가진 존재일 수 있는가? 자율성은 책임과 권리같은 관념의 선결 조건으로 오늘의 사회에서 ‘인공지능이 해내는 일’에 연관된 책임과 공헌을 배분하는 데 적용할 규범의 이론적 기반이 된다는 점에서 이 물음은 현실적으로 중요하다. 공학의 ‘자율적 에이전트’ 개념은 현행의 철학적 논의에도 영향을 미치며, 공학과 철학의 자율성 개념 간의 관계가 명료하게 인식되지 못한 까닭에 야기될 혼란의 개연성이 실재한다. 이런 현실의 문제 상황을 의식하면서, 이 논문은 앞의 물음을 가능성에 관한 물음과 정당성에 관한 물음으로 분석하고, 인공지능의 자율성에 관한 전망을 검토한다. 논의의 과정에서 ‘X가 자율성을 가진다’는 (...) 의미를 규명해야 할 필요가 확인되고, 논문은 루소와 칸트와 크리스만의 논의를 활용하여 자율성의 개념을 고찰한다. 그렇게 분석한 자율성의 개념을 인공지능의 속성에 적용해 볼 때, 칸트가 정립한 자율성 개념에 비추어 인공지능은 자율적 존재일 수 없으며, 크리스만의 서술에서 부각된 자율적 존재의 핵심 속성을 고려할 때 인공지능이나 그것으로 작동하는 로보틱 시스템이 자율성을 지니도록 하는 것은 부적절한 위험을 허용한다는 점에서 옳지 않음이 드러난다. (shrink)

Applications of learning to autonomous agents (simulated or real) have often been restricted to learning a mapping from perceived state of the world to the next action to take. Often this is couched in terms of learning from no previous knowledge. This general case for real autonomous robots is very difficult. In any case, when building a real robot there is usually a lot of a priori knowledge (e.g., from the engineering that went into its design) which doesn’t need (...) to be learned. We describe the behavior-based approach to autonomous robots, and then examine four classes of learning problems associated with such robots. (shrink)

The two-volume set LNCS 10350 and 10351 constitutes the thoroughly refereed proceedings of the 30th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems, IEA/AIE 2017, held in Arras, France, in June 2017. The 70 revised full papers presented together with 45 short papers and 3 invited talks were carefully reviewed and selected from 180 submissions. They are organized in topical sections: constraints, planning, and optimization; data mining and machine learning; sensors, signal processing, and data fusion; (...) recommender systems; decision support systems; knowledge representation and reasoning; navigation, control, and autonome agents; sentiment analysis and social media; games, computer vision; and animation; uncertainty management; graphical models: from theory to applications; anomaly detection; agronomy and artificial intelligence; applications of argumentation; intelligent systems in healthcare and mhealth for health outcomes; and innovative applications of textual analysis based on AI. (shrink)

In this paper I provide an exposition and critique of Johnson and Noorman’s (2014) three conceptualizations of the agential roles artificial systems can play. I argue that two of these conceptions are unproblematic: that of causally efficacious agency and “acting for” or surrogate agency. Their third conception, that of “autonomous agency,” however, is one I have reservations about. The authors point out that there are two ways in which the term “autonomy” can be used: there is, firstly, the engineering (...) sense of the term, which simply refers to the ability of some system to act independently of substantive human control. Secondly, there is the moral sense of the term, which has traditionally grounded many notions of human mor-al responsibility. I argue that the continued usage of “autonomy” in discussions of artificial agency complicates matters unnecessarily. This occurs in two ways: firstly, the condition of autonomy, even in its engineering sense, fails to accurately describe the way “autonomous” systems are developed in practice. Secondly, the continued usage of autonomy in the moral sense introduces unnecessary meta-physical baggage form the free will debate into discussions about moral agency. In order to understand the debate surrounding autonomy, we would therefore first need to settle many seemingly intractable metaphysical questions regarding the existence of free will in human beings. (shrink)

what is now the mainstream view as to the best way forward in the dream of engineering reliable software systems out of autonomous agents. The way of using formal logics to specify, implement and verify distributed systems of interacting units using a guiding analogy of beliefs, desires and intentions. The implicit message behind the book is this: Distributed Artificial Intelligence (DAI) can be a respectable engineering science. It says: we use sound formal systems; can cite established philosophical foundations; (...) and will be able to build reliable and flexible software systems. (shrink)

This book is an archetypal product of the Belief-Desire-Intention (BDI) school of multi-agent systems. It presents what is now the mainstream view as to the best way forward in the dream of engineering reliable software systems out of autonomous agents. The way of using formal logics to specify, implement and verify distributed systems of interacting units using a guiding analogy of beliefs, desires and intentions. The implicit message behind the book is this: Distributed Artificial Intelligence (DAI) can be a (...) respectable engineering science. It says: we use sound formal systems; can cite established philosophical foundations; and will be able to build reliable and flexible software systems. (shrink)

One of the primary, if not most critical, difficulties in the design and implementation of autonomous systems is the black-boxed nature of the decision-making structures and logical pathways. How human values are embodied and actualised in situ may ultimately prove to be harmful if not outright recalcitrant. For this reason, the values of stakeholders become of particular significance given the risks posed by opaque structures of intelligent agents (IAs). This paper explores how decision matrix algorithms, via the belief-desire-intention model for (...) autonomous vehicles, can be designed to minimize the risks of opaque architectures. Primarily through an explicit orientation towards designing for the values of explainability and verifiability. In doing so, this research adopts the Value Sensitive Design (VSD) approach as a principled framework for the incorporation of such values within design. VSD is recognized as a potential starting point that offers a systematic way for engineering teams to formally incorporate existing technical solutions within ethical design, while simultaneously remaining pliable to emerging issues and needs. It is concluded that the VSD methodology offers at least a strong enough foundation from which designers can begin to anticipate design needs and formulate salient design flows that can be adapted to the changing ethical landscapes required for utilisation in autonomous vehicles. (shrink)

The imminent deployment of autonomous vehicles requires algorithms capable of making moral decisions in relevant traffic situations. Some scholars in the ethics of autonomous vehicles hope to align such intelligent systems with human moral judgment. For this purpose, studies like the Moral Machine Experiment have collected data about human decision-making in trolley-like traffic dilemmas. This paper first argues that the trolley dilemma is an inadequate experimental paradigm for investigating traffic moral judgments because it does not include agents’ character-based considerations and (...) is incapable of facilitating the investigation of low-stakes mundane traffic scenarios. In light of the limitations of the trolley paradigm, this paper presents an alternative experimental framework that addresses these issues. The proposed solution combines the creation of mundane traffic moral scenarios using virtual reality and the Agent-Deed-Consequences (ADC) model of moral judgment as a moral-psychological framework. This paradigm shift potentially increases the ecological validity of future studies by providing more realism and incorporating character considerations into traffic actions. (shrink)

In recent years, autonomous systems have become an important research area and application domain, with a significant impact on modern society. Such systems are characterized by different levels of autonomy and complex communication infrastructures that allow for collective decision-making strategies. There exist several publications that tackle ethical aspects in such systems, but mostly from the perspective of a single agent. In this paper we go one step further and discuss these ethical challenges from the perspective of an aggregate of autonomous (...) systems capable of collective decision-making. In particular, in this paper, we propose the Caesar approach through which we model the collective ethical decision-making process of a group of actors—agents and humans, as well as define the building blocks for the agents participating in such a process, namely Caesar agents. Factors such as trust, security, safety, and privacy, which affect the degree to which a collective decision is ethical, are explicitly captured in Caesar. Finally, we argue that modeling the collective decision-making in Caesar provides support for accountability. (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)

This article proposes a number of models to examine through which mechanisms a population of autonomous agents could arrive at a repertoire of perceptually grounded categories that is sufficiently shared to allow successful communication. The models are inspired by the main approaches to human categorisation being discussed in the literature: nativism, empiricism, and culturalism. Colour is taken as a case study. Although we take no stance on which position is to be accepted as final truth with respect to human categorisation (...) and naming, we do point to theoretical constraints that make each position more or less likely and we make clear suggestions on what the best engineering solution would be. Specifically, we argue that the collective choice of a shared repertoire must integrate multiple constraints, including constraints coming from communication. Key Words: autonomous agents; colour categorisation; colour naming; connectionism; cultural evolution; genetic evolution; memes; origins of language; self-organisation; semiotic dynamics; symbol grounding. (shrink)

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

In the fields of artificial intelligence and robotics, the term “autonomy” is generally used to mean the capacity of an artificial agent to operate independently of human guidance. It is thereby assumed that the agent has a fixed goal or “utility function” with respect to which the appropriateness of its actions will be evaluated. From a philosophical perspective, this notion of autonomy seems oddly weak. For, in philosophy, the term is generally used to refer to a stronger capacity, namely the (...) capacity to “give oneself the law,” to decide by oneself what one’s goal or principle of action will be. The predominant view in the literature on the long-term prospects and risks of artificial intelligence is that an artificial agent cannot exhibit such autonomy because it cannot rationally change its own final goal, since changing the final goal is counterproductive with respect to that goal and hence undesirable. The aim of this paper is to challenge this view by showing that it is based on questionable assumptions about the nature of goals and values. I argue that a general AI may very well come to modify its final goal in the course of developing its understanding of the world. This has important implications for how we are to assess the long-term prospects and risks of artificial intelligence. (shrink)

Information and representation are thought to be intimately related. Representation, in fact, is commonly considered to be a special kind of information. It must be a _special_ kind, because otherwise all of the myriad instances of informational relationships in the universe would be representational -- some restrictions must be placed on informational relationships in order to refine the vast set into those that are truly representational. I will argue that information in this general sense is important to genuine agents, but (...) that it is a blind alley with regard to the attempt to understand representation. On the other hand, I will also argue that a different, quite non-standard, form of information is central to genuine representation. First I turn to some of the reasons why information as usually considered is the wrong category for understanding representation; second to an alternative model of representation -- one that is naturally emergent in autonomous agents, and that does involve information, but not in any standard form; and third I return to standard notions of informational relationships and show what they are in fact useful for. (shrink)

The phenomenological goal of grounding the content of conceptual thought in the background understanding of everyday, skillful coping was approached using evolutionary autonomous agent methodology. The behavior of an EAA evolved to perform a specified motor task was identified with skillful coping. Changes in the dynamics of the EAA controller occurred when the EAA encountered an unexpected obstacle with loss of longer time scale components in its hierarchical temporal organization. These temporal changes are consistent with the phenomenological changes which we (...) experience with breakdown during equipment use with our adoption of a more immediate, determinate stance. Since this latter experience is the basis of conceptual thought, the EAA paradigm goes some way in providing a naturalized explanation for the grounding of the content of conceptual thought in everyday, skillful coping in a manner that is physiologically plausible and phenomenologically accurate. (shrink)

Human organizations’ adoption of the paradigm of the Fourth Industrial Revolution is associated with the growth of techno-empowerment, which is the process of transferring autonomy in decision-making to intelligent machines. Particular persuasive strategies have been identified that may coax people to use intelligent devices. However, there is a substantial research gap regarding what antecedents influence human intention to assign decision-making autonomy to artificial agents. In this study, ethological and evolutionary concepts are applied to explain the drivers for autonomous assistants’ techno-empowerment. (...) The method used in the study was a 4 × 2 between-subject experiment made with 278 persons. The research tool used to collect the data was an online survey. The results show that more positive attitudes and higher trust, perceived usefulness, and perceived ease of use are correlated with higher intention to allow the autonomous assistant independence in decision-making. Second, the results suggest that the more human-like a non-human agent is, the higher the intention to empower it—but only if this agent simultaneously provides functional and visual anthropomorphic cues explainable by the mimicry effect. (shrink)

In this paper, I would like to show that considering technological models as they arise in engineering disciplines can greatly enrich the philosophical perspective on models. In fluid mechanics, (at least) three types of models are distinguished: mathematical, computer and physical models. Very often, the choice of a particular mathematical, computer or physical model highly affects the type of solutions and the computational time needed for it. Technological models not only aim at a correct description of the physical phenomena, (...) but also for an efficient and accurate simulation. The problem arises how heterogeneous models of an engineering problem can be brought together and be compared to each other as regards their function and technological efficiency. There are two developments in the history of fluid mechanics that have greatly influenced the use of models in the field: The introduction of the concept of the boundary layer by Ludwig Prandtl in 1904 made it possible to apply ideal analytical solutions, which at the time were almost entirely based on Euler’s equations for inviscid fluids, to interesting real cases and to approximate the theoretical Navier-Stokes equations to practical engineering problems, i.e. to cases at high Reynolds numbers. This made it possible to link the empirical tradition of hydraulics with the theoretical tradition of analytical mechanics and therefore lead to a kind of equilibrium in the use of mathematical and physical models. In the 1970s the introduction of the computer has greatly pushed back the importance of both physical and mathematical (analytic) models alike without making them superfluous. There remain, however, three different ways to conceive of physical models in fluid mechanics, and thus of the experimental ingredient, depending on whether they are devised from an analytical, computational or measurement theoretical point of view. Yet even inside the tradition of computer simulation, different practices have formed according to the programming methods used. The choice of method either depends on considerations of efficiency in terms of costs and time, or on historically contingent factors, like availability of instruments and programming packages or the arbitrary choice of a forerunner. Seen from a technological point of view the factors that make models “autonomous agents” and thus (relatively) independent from theory depend on efficiency constraints. Models are means to solve problems in a certain practical perspective by the most efficient means available. To develop a model is a “fast and frugal way” to get to grips with a certain region of reality, whereas the theoretical approach stresses the importance of universal features. (shrink)

The control principle implies that people should not feel guilt for outcomes beyond their control. Yet, the so-called ‘agent and observer puzzles’ in philosophy demonstrate that people waver in their commitment to the control principle when reflecting on accidental outcomes. In the context of car accidents involving conventional or autonomous vehicles, Study 1 established that judgments of responsibility are most strongly associated with expressions of guilt–over and above other negative emotions, such as sadness, remorse or anger. Studies 2 and 3 (...) then confirmed that, while people generally endorse the control principle, and deny that occupants in an AV should feel guilt when involved in an accident, they nevertheless ascribe guilt to those same occupants. Study 3 also uncovered novel implications of the observer puzzle in the legal context: Passengers in an AV were seen as more legally liable than either passengers in a conventional vehicle, or even their drivers–especially when participants were prompted to reflect on the passengers’ affective experience of guilt. Our findings document an important conflict–in the context of AV accidents–between people’s prescriptive reasoning about responsibility and guilt on one hand, and their counter-normative experience of guilt on the other, with apparent implications for liability decisions. (shrink)

The design of collaborative robotics, such as driver-assisted operations, engineer a potential automation of decision-making predicated on unobtrusive data gathering of human users. This form of ‘somatic surveillance’ increasingly relies on behavioural biometrics and sensory algorithms to verify the physiology of bodies in cabin interiors. Such processes secure cyber-physical space, but also register user capabilities for control that yield data as insured risk. In this technical re-formation of human–machine interactions for control and communication ‘a dissonance of attribution’ :7684, 2019. https://doi.org/10.1073/pnas.1805770115) (...) is created between perceptions of phenomena, materials and decision-making. This reconfigures relations not only between humans and machines, objects and subjects, but possibly disrupts attributive functions in the social system of Law. What it requires is shifting a legal accountability for action from a sovereignty of the human to a new materialist account based on a ‘cognitive assemblage’ between physiological data, computation and algorithmic sensing. This paper investigates the function of law as a guidance system to acknowledge this account of sensory and algorithmic computation as autonomous ‘sensing agents’ that may be accountable in situations of risk. This assemblage of robotic computation and sensory determination requires a clearer legal differentiation across the current static terminologies of person, property, liability and rights that maintain strict separations of object from subject. To neglect this, we argue, law will solely impute attributions of error to humans despite evidence of operation via mutual control. (shrink)

In this paper we present some concepts and their relations that are necessary for modeling autonomous agents in an environment that is governed by some (social) norms. We divide the norms over three levels: the private level the contract level and the convention level. We show how deontic logic can be used to model the concepts and how the theory of speech acts can be used to model the generation of (some of) the norms. Finally we give some idea about (...) an agent architecture incorporating the social norms based on a BDI framework. (shrink)

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

Children continue to be overlooked as a topic of concern in discussions around the ethical use of people’s data and information. Where children are the subject of such discussions, the focus is often primarily on privacy concerns and consent relating to the use of their data. This paper highlights the unique challenges children face when it comes to online interferences with their decision-making, primarily due to their vulnerability, impressionability, the increased likelihood of disclosing personal information online, and their developmental capacities. (...) These traits allow for practices such as hypernudging to be executed on them more accurately and with more serious consequences, specifically by potentially undermining their autonomy. We argue that children are autonomous agents in the making and thus require additional special protections to ensure that the development of their autonomy is safeguarded. This means that measures should be taken to prohibit most forms of hypernudging children and thus ensure that they are protected from this powerful technique of digital manipulation. (shrink)

Legal structures may form barriers to, or enablers of, adoption of precision agriculture management with small autonomous agricultural robots. This article develops a conceptual regulatory framework for small autonomous agricultural robots, from a practical, self-contained engineering guide perspective, sufficient to get working research and commercial agricultural roboticists quickly and easily up and running within the law. The article examines the liability framework, or rather lack of it, for agricultural robotics in EU, and their transpositions to UK law, as a (...) case study illustrating general international legal concepts and issues. It examines how the law may provide mitigating effects on the liability regime, and how contracts can be developed between agents within it to enable smooth operation. It covers other legal aspects of operation such as the use of shared communications resources and privacy in the reuse of robot-collected data. Where there are some grey areas in current law, it argues that new proposals could be developed to reform these to promote further innovation and investment in agricultural robots. (shrink)

As robots and Artificial Intelligences become more enmeshed in rich social contexts, it seems inevitable that we will have to make them into moral machines equipped with moral skills. Apart from the technical difficulties of how we could achieve this goal, we can also ask the ethical question of whether we should seek to create such Artificial Moral Agents (AMAs). Recently, several papers have argued that we have strong reasons not to develop AMAs. In response, we develop a comprehensive analysis (...) of the relevant arguments for and against creating AMAs, and we argue that all things considered we have strong reasons to continue to responsibly develop AMAs. The key contributions of this paper are threefold. First, to provide the first comprehensive response to the important arguments made against AMAs by Wynsberghe and Robbins (in “Critiquing the Reasons for Making Artificial Moral Agents”, Science and Engineering Ethics 25, 2019) and to introduce several novel lines of argument in the process. Second, to collate and thematise for the first time the key arguments for and against AMAs in a single paper. Third, to recast the debate away from blanket arguments for or against AMAs in general, to a more nuanced discussion about the use of what sort of AMAs, in what sort of contexts, and for what sort of purposes is morally appropriate. (shrink)

It is an increasingly popular view among philosophers that moral responsibility can, in principle, be attributed to unconscious autonomous agents. This trend is already remarkable in itself, but it is even more interesting that most proponents of this view provide more or less the same argument to support their position. I argue that as it stands, the Extension Argument, as I call it, is not sufficient to establish the thesis that unconscious autonomous agents can be morally responsible. I attempt to (...) show that the Extension Argument should overcome especially strong ethical considerations; moreover, its epistemological grounds are not too solid, partly because the justifications of its premises are in conflict. (shrink)

Alfred Mele’s Autonomous Agents offers a penetrating treatment of autonomy. Understood as an actual condition of self-rule, autonomy is nested within the range of freedom concepts often associated with discussions of moral responsibility. In part 1 of his two-part Autonomous Agents, Mele attempts to capture autonomy by exploring the upper reaches of self-control, where self-control is understood as the opposite of akrasia, that is, weakness of will. It is Mele’s contention that even an optimally self-controlled agent could fall short of (...) autonomy. What, he asks, can be added to an ideally self-controlled agent to provide conditions sufficient for autonomous agency? In part 2, Mele offers not one answer but two. He offers first an account that would satisfy a compatibilist, and then one that would meet the further demands of an incompatibilist. Mele himself remains agnostic as regards the debate between compatibilists and incompatibilists. He does, however, shape novel accounts of each by exploiting the resources that, he argues, are needed to move past self-control to autonomy. He closes by arguing that we have better reason to believe that there are autonomous agents than we do to believe that there are none. (shrink)

A body schema is an agent's model of its own body that enables it to act on affordances in the environment. This paper presents a body schema system for the Learning Intelligent Decision Agent (LIDA) cognitive architecture. LIDA is a conceptual and computational implementation of Global Workspace Theory, also integrating other theories from neuroscience and psychology. This paper contends that the ‘body schema' should be split into three separate functions based on the functional role of consciousness in Global Workspace (...) Theory. There is (1) an online model of the agent's effectors and effector variables (Current Body Schema), (2) a long-term, recognitional storage of embodied capacities for action and affordances (Habitual Body Schema), and (3) "dorsal" stream information feeding directly from early perception to sensorimotor processes (Online Body Schema). This paper then discusses how the LIDA model of the body schema explains several experiments in psychology and ethology. (shrink)

Agents and agent-based systems are becoming essential in the development of various fields, such as artificial intelligence, ubiquitous computing, ambient intelligence, autonomous computing, and intelligent robotics. The concept of autonomous agents, inspired by the observed agency in living systems, is also central to current theories on the origin, development, and evolution of life. Therefore, it is crucial to develop an accurate understanding of agents and the concept of agency. This paper begins by discussing the role of agency in natural systems (...) as an inspiration and motivation for agential technologies and then introduces the idea of artificial agents. A systematic approach is presented for the classification of artificial agents. This classification aids in understanding the existing state of the artificial agents and projects their potential future roles in addressing specific types of problems with dedicated agent types. (shrink)

Living-donor kidney transplantation is the “gold standard” treatment for many individuals with end-stage renal failure. Superior outcomes for the graft and the transplant recipient have prompted the implementation of new strategies promoting living-donor kidney transplantation, and the number of such transplants has increased considerably over recent years. Living donors are undoubtedly exposed to risk. In his editorial “underestimating the risk in living kidney donation”, Walter Glannon suggests that more data on long-term outcomes for living donors are needed to determine whether (...) this risk is permissible and the extent to which physicians and transplant surgeons should promote living-donor kidney transplantation.1 In this paper I argue that it is not clear that medical professionals have underestimated this risk, nor is it clear that more data on long-term outcomes are needed in order to determine whether it is permissible for individual autonomous agents to expose themselves to this or, indeed, any risk. The global shortage of organs available for transplantation ultimately means that every year thousands of individuals who value their life die needlessly. This is an unacceptable loss of human life. Saving life is one of the most wonderful things an individual can do for another. Promoting any strategy that will assist in saving life and preventing human suffering within acceptable moral limits is legitimate. (shrink)

Interactions between an intelligent software agent and a human user are ubiquitous in everyday situations such as access to information, entertainment, and purchases. In such interactions, the ISA mediates the user’s access to the content, or controls some other aspect of the user experience, and is not designed to be neutral about outcomes of user choices. Like human users, ISAs are driven by goals, make autonomous decisions, and can learn from experience. Using ideas from bounded rationality, we frame these interactions (...) as instances of an ISA whose reward depends on actions performed by the user. Such agents benefit by steering the user’s behaviour towards outcomes that maximise the ISA’s utility, which may or may not be aligned with that of the user. Video games, news recommendation aggregation engines, and fitness trackers can all be instances of this general case. Our analysis facilitates distinguishing various subcases of interaction, as well as second-order effects that might include the possibility for adaptive interfaces to induce behavioural addiction, and/or change in user belief. We present these types of interaction within a conceptual framework, and review current examples of persuasive technologies and the issues that arise from their use. We argue that the nature of the feedback commonly used by learning agents to update their models and subsequent decisions could steer the behaviour of human users away from what benefits them, and in a direction that can undermine autonomy and cause further disparity between actions and goals as exemplified by addictive and compulsive behaviour. We discuss some of the ethical, social and legal implications of this technology and argue that it can sometimes exploit and reinforce weaknesses in human beings. (shrink)

Alasdair Cochrane introduces an entirely new theory of animal rights grounded in their interests as sentient beings. He then applies this theory to different and underexplored policy areas, such as genetic engineering, pet-keeping, indigenous hunting, and religious slaughter. In contrast to other proponents of animal rights, Cochrane claims that because most sentient animals are not autonomous agents, they have no intrinsic interest in liberty. As such, he argues that our obligations to animals lie in ending practices that cause their (...) suffering and death and do not require the liberation of animals. Cochrane's "interest-based rights approach" weighs the interests of animals to determine which is sufficient to impose strict duties on humans. In so doing, Cochrane acknowledges that sentient animals have a clear and discernable right not to be made to suffer and not to be killed, but he argues that they do not have a prima facie right to liberty. Because most animals possess no interest in leading freely chosen lives, humans have no moral obligation to liberate them. Moving beyond theory to the practical aspects of applied ethics, this pragmatic volume provides much-needed perspective on the realities and responsibilities of the human-animal relationship. (shrink)

The transfer of tasks with sometimes far-reaching implications to autonomous systems raises a number of ethical questions. In addition to fundamental questions about the moral agency of these systems, behavioral issues arise. We investigate the empirically accessible question of whether the imposition of harm by an agent is systematically judged differently when the agent is artificial and not human. The results of a laboratory experiment suggest that decision-makers can actually avoid punishment more easily by delegating to machines than by delegating (...) to other people. Our results imply that the availability of artificial agents could provide stronger incentives for decision-makers to delegate sensitive decisions. (shrink)

Cognitive science's basic premises are under attack. In particular, its focus on internal cognitive processes is a target. Intelligence is increasingly interpreted, not as a matter of reclusive thought, but as successful agent-environment interaction. The critics claim that a major reorientation of the field is necessary. However, this will only occur when there is a distinct alternative conceptual framework to replace the old one. Whether or not a serious alternative is provided is not clear. Among the critics there is some (...) consensus, however, that this role could be fulfilled by the concept of a 'behavioral system'. This integrates agent and environment into one encompassing general system. We will discuss two contexts in which the behavioral systems idea is being developed. Autonomous Agents Research is the enterprise of building behavior-based robots. Dynamical Systems Theory provides a mathematical framework well suited for describing the interactions between complex systems. We will conclude that both enterprises provide important contributions to the behavioral systems idea. But neither turns it into a full conceptual alternative which will initiate a major paradigm switch in cognitive science. The concept will need a lot of fleshing out before it can assume that role. (shrink)

We examine Van Wynsberghe and Robbins (JAMA 25:719-735, 2019) critique of the need for Artificial Moral Agents (AMAs) and its rebuttal by Formosa and Ryan (JAMA 10.1007/s00146-020-01089-6, 2020) set against a neo-Aristotelian ethical background. Neither Van Wynsberghe and Robbins (JAMA 25:719-735, 2019) essay nor Formosa and Ryan’s (JAMA 10.1007/s00146-020-01089-6, 2020) is explicitly framed within the teachings of a specific ethical school. The former appeals to the lack of “both empirical and intuitive support” (Van Wynsberghe and Robbins 2019, p. 721) for (...) AMAs, and the latter opts for “argumentative breadth over depth”, meaning to provide “the essential groundwork for making an all things considered judgment regarding the moral case for building AMAs” (Formosa and Ryan 2019, pp. 1–2). Although this strategy may benefit their acceptability, it may also detract from their ethical rootedness, coherence, and persuasiveness, characteristics often associated with consolidated ethical traditions. Neo-Aristotelian ethics, backed by a distinctive philosophical anthropology and worldview, is summoned to fill this gap as a standard to test these two opposing claims. It provides a substantive account of moral agency through the theory of voluntary action; it explains how voluntary action is tied to intelligent and autonomous human life; and it distinguishes machine operations from voluntary actions through the categories of poiesis and praxis respectively. This standpoint reveals that while Van Wynsberghe and Robbins may be right in rejecting the need for AMAs, there are deeper, more fundamental reasons. In addition, despite disagreeing with Formosa and Ryan’s defense of AMAs, their call for a more nuanced and context-dependent approach, similar to neo-Aristotelian practical wisdom, becomes expedient. (shrink)

This paper investigates decision-theoretic planning in sophisticated autonomous agents operating in environments of real-world complexity. An example might be a planetary rover exploring a largely unknown planet. It is argued th a t existing algorithms for decision-theoretic planning are based on a logically incorrect theory of rational decision making. Plans cannot be evaluated directly in terms of their expected values, because plans can be of different scopes, and they can interact with other previously adopted plans. Furthermore, in the real world, (...) the search for optimal plans is completely intractable. An alternative theory of rational decision making is proposed, called “locally global planning”. (shrink)

Computational approaches based on autonomous agents share with new ape language research the same principles of dynamical system paradigms. A recent model for the evolution of symbolization and language in autonomous agents is briefly described in order to highlight the similarities between these two methodologies. The additional benefits of autonomous agent modeling in the field of language origin research are highlighted.

A socio-cognitive approach to trust can help us envisage a notion of networked trust for multi-agent systems (MAS) based on different interacting agents. In this framework, the issue is to evaluate whether or not a socio-cognitive analysis of trust can apply to the interactions between human and autonomous agents. Two main arguments support two alternative hypothesis; one suggests that only reliance applies to artificial agents, because predictability of agents’ digital interaction is viewed as an absolute value and human relation is (...) judged to be a necessary requirement for trust. The other suggests that trust may apply to autonomous agents because predictability of agents’ interaction is viewed only as a relative value since the digital normativity that grows out of the communication process between interacting agents in MAS has always deal with some unpredictable outcomes (_reduction of uncertainty_). Furthermore, human touch is not judged to be a necessary requirement for trust. In this perspective, a diverse notion of trust is elaborated, as trust is no longer conceived only as a relation between interacting agents but, rather, as a relation between cognitive states of control and lack of control (_double bind_). (shrink)

Many open networks, distributed computing systems, and infrastructure management systems face a common problem: how to distribute a collectivised set of resources amongst a set of autonomous agents of heterogenous provenance. One approach is for the agents themselves to self-organise the allocation of resources with respect to a set of agreed conventional rules; but given an allocation scheme which maps resources to those agents and a set of rules for determining that allocation scheme, some natural questions arise—Is this allocation fair? (...) Is the allocation method effective? Is it efficient? Are the decision makers accountable? In this paper, we argue that some answers to these questions can be found in the formal characterisation of different aspects of ‘justice’ and that these different aspects need a principled operationalisation as policies for system management. We present a formal model and some experimental results, concluding that the different aspects are all inter-connected and that what is required is a comprehensive research programme in computational justice. (shrink)

Across various fields it is argued that the self in part consists of an autobiographical self-narrative and that the self-narrative has an impact on agential behavior. Similarly, within action theory, it is claimed that the intentional structure of coherent long-term action is divided into a hierarchy of distal, proximal, and motor intentions. However, the concrete mechanisms for how narratives and distal intentions are generated and impact action is rarely fleshed out concretely. We here demonstrate how narratives and distal intentions can (...) be generated within cognitive agents and how they can impact agential behavior over long time scales. We integrate narratives and distal intentions into the LIDA model,and demonstrate how they can guide agential action in a manner that is consistent with the Global Workspace Theory of consciousness. This paper serves both as an addition to the LIDA cognitive architecture and an elucidation of how narratives and distal intention emerge and play their role in cognition and action. (shrink)

The recent progress in the development of autonomous cars has seen ethical questions come to the forefront. In particular, life and death decisions regarding the behavior of self-driving cars in trolley dilemma situations are attracting widespread interest in the recent debate. In this essay we want to ask whether we should implement a mandatory ethics setting for the whole of society or, whether every driver should have the choice to select his own personal ethics setting. While the consensus view seems (...) to be that people would not be willing to use an automated car that might sacrifice themselves in a dilemma situation, we will defend the somewhat contra-intuitive claim that this would be nevertheless in their best interest. The reason is, simply put, that a PES regime would most likely result in a prisoner’s dilemma. (shrink)

This paper addresses the logical foundations of goal-regression planning in autonomous rational agents. It focuses mainly on three problems. The first is that goals and subgoals will often be conjunctions, and to apply goal-regression planning to a conjunction we usually have to plan separately for the conjuncts and then combine the resulting subplans. A logical problem arises from the fact that the subplans may destructively interfere with each other. This problem has been partially solved in the AI literature (e.g., in (...) SNLP and UCPOP), but the solutions proposed there work only when a restrictive assumption is satisfied. This assumption pertains to the computability of threats. It is argued that this assumption may fail for an autonomous rational agent operating in a complex environment. Relaxing this assumption leads to a theory of defeasible planning. The theory is formulated precisely and an implementation in the OSCAR architecture is discussed. (shrink)

Mele questions whether being a self-controlled person is also sufficient for personal autonomy. He constructs an ideally self-controlled person and argues that such a person may lack autonomy in certain ways. The task, then, is to determine what needs to be added to the ideally self-controlled person in order to make him autonomous. Throughout the book, Mele is concerned with responding to objections from both compatibilist and incompatibilist philosophers.

Mele endeavors to analyze autonomy, understood as an actual condition of self-governance, in two stages. An account of ideally self-controlled agents is developed first. This account is then supplemented with features sufficient for individual autonomy.

We present a logic-based formalism for modeling ofdialogues between intelligent and autonomous software agents,building on a theory of abstract dialogue games which we present.The formalism enables representation of complex dialogues assequences of moves in a combination of dialogue games, and allowsdialogues to be embedded inside one another. The formalism iscomputational and its modular nature enables different types ofdialogues to be represented.