Machine Ethics

Utilitarian concerns increasingly drive machine ethics, particularly the few ethical discussions in learning analytics. Determining the most good for the most people is a responsible way forward, but perhaps not the most comprehensive, especially when assessing potential outcomes. Innovations in learning analytics may occur more rapidly and with better outcomes if informed ethical discussions occur at every step of development.

David J. Gunkel was a keynote speaker for “Identity, Agency, and the Digital Nexus”, April 2013, SSHRC funded  international symposium hosted by Athabasca University. His talk challenged the audience to reframe and rethink the “human-machine” binary in 21st century understandings of ethics and agency. Later that year Gunkel talked with Paul Kellogg over Skype to discuss some of the points raised in the presentation, and other key ideas embedded in Gunkel's 2012 publication, "The Machine Question: Critical Perspectives on AI, Robots and Ethics." Gunkel is the Presidential Teaching Professor of Communication Studies at Northern Illinois University. The symposium was funded by a Connection Grant from the Social Sciences and Humanities Research Council of Canada and organized by Dr. Raphael Foshay, MA-IS Program Director at Athabasca University.

Robot surgery began as a general practice in the
United States in 2000 when the robotic da Vinci Surgical
System was approved for use in hospitals by the FDA.
Recently there have been some questions regarding the
safety record of these machines and their benefits relative
to their expense. Should we as patients and caregivers place
more or less trust in the technology of robotic surgery? To
answer this question we will look at the possibility that
robotic surgery may be an example of reverse adaptation,
where the technology drives the social contract between
the doctor and patient. Additionally, we will look at the
impacts robotic surgery will have on traditional aspects of
the medical ethics of surgery, such as informed consent,
autonomy of surgeons and patients, corporate marketing,
and the duty to provide the best available care, as well as
increasing the asymmetry of the trust relationship between
surgeon and patient. It will be argued that these issues
will grow in importance as future robot surgery systems
gain more autonomy in making or suggesting surgical
strategies.

Social robots are those endowed with communication channels and abilities that take inspiration from human beings. The scope of such abilities should include those allowing a robot to understand people’s affective states and expressions, intentions, actions, and to interpret them based on contextual information. Childcare robots are an example of robots that could take advantage of the integration of these capabilities. This commentary conducts a technical appraisal of the notion of autonomous childcare robots, focusing on these social perceptive capabilities, reviewing some of the key challenges remaining to be investigated by the research community in this respect.

This presentation was delivered at the First Annual Conference on the Governance of Emerging Technology organized by the Sandra Day O'Connor College of Law at Arizona State University. The presentation makes a case for extending rights to artificial autonomous agents, AIs, and robots. It is based on research published in "The Machine Question: Critical Perspectives on AI, Robots, and Ethics" (MIT 2012).

Building simulated affects into artifacts poses a moral dilemma. On the one hand, in order for humans to interact fully with machines, the machines need to meet them on human terms, and this requires machines that are capable of assessing human affective states and responding to them in kind. On the other hand, doing so amounts to a fundamental deception that humans will find it hard to keep in mind, namely that these machines do not actually have these affective states and may not, therefore, be worthy of our attachment and our moral regard. Yet, simulated affects, we argue, are necessary for creating machines that can make moral decisions, that affects are essential for disambiguating the utterances of machines and humans, and that not all forms of deception are bad. After an initial provocation, these arguments are taken up in order, after which we argue further that while simulated affects are necessary, they could nonetheless lead to abuse. Therefore, the standard ethical limits and principles involved with any technological innovation must be respected.

Stephen Hawking recently told the BBC that "the development of full artificial intelligence (AI) could spell the end of the human race." The real problem here is not with the rather disturbing predictions of Hawking, but with the apocalyptic tone and mode of thinking that he employs and utilizes.

The development of robotics and AI agents has enabled their wider usage in human surroundings. AI agents are more trusted to make increasingly important decisions with potentially critical outcomes. It is essential to consider the ethical consequences of the decisions made by these systems. In this paper, we present how contrastive explanations can be used for comparing the ethics of plans. We build upon an existing ethical framework to allow users to make suggestions to plans and receive contrastive explanations.

In this paper I argue that in certain circumstances robots can be seen as real moral agents. A distinction is made between persons and moral agents such that, it is not necessary for a robot to have personhood in order to be a moral agent. I detail three requirements for a robot to be seen as a moral agent. The first is achieved when the robot is significantly autonomous from any programmers or operators of the machine. The second is when one can analyze or explain the robot's behavior only by ascribing to it some predisposition or 'intention' to do good or harm. And finally, robot moral agency requires the robot to behave in a way that shows and understanding of responsibility to some other moral agent. Robots with all of these criteria will have moral rights as well as responsibilities regardless of their status as persons.

Etica dell'Intelligenza Artificiale: una FUTUROMACHÍA? Intelligenza artificiale, etica e futuro In un'epoca tanto definita dalle conquiste della scienza tecnolo-gica, la sfida che il futuro riserva al pensiero morale non può che presentarsi anche come riflessione critica sulla relazione che intercorre tra tecnologie digitali e vita umana. In particolare, l'etica dell'Intelli-genza Artificiale (IA) ha conosciuto negli ultimi decenni una notevole affermazione, 1 tanto da essere divenuta ingrediente fondamentale di ogni discussione circa lo sviluppo futuro della tecnologia e la sua in-tegrazione alla società. 2 Per quanto l'IA stia già riconfigurando vari settori del mondo con-temporaneo, è soprattutto la dimensione futura del suo sviluppo ed utilizzo a imporre l'urgenza di un suo vaglio in chiave morale. L'etica dell'IA-il cui compito, in sintesi, consiste sia nella critica dei suoi impatti morali sia nell'elaborazione di princìpi, procedure e racco-mandazioni per la sua progettazione e sviluppo etico-è legata alla dimensione futura in almeno tre nodi.

GUGLIELMO TAMBURRINI - ETICA DELLE MACCHINE. DILEMMI MORALI PER ROBOTICA E INTELLIGENZA ARTIFICIALE

CAROCCI EDITORE - 152 pp.

Collana Quality paperbacks - ISBN: 9788843098965

http://www.carocci.it/index.php?option=com_carocci&task=schedalibro&Itemid=72&isbn=9788843098965

INDICE

INTRODUZIONE

1. REGOLE MORALI PER VEICOLI AUTONOMI
L’ingegnere e l’etica delle conseguenze/Sicurezza stradale e veicoli autonomi/Libertà personale ed etica delle conseguenze/Dilemmi morali nelle collisioni inevitabili/Tensioni nell’etica delle conseguenze/Tensioni tra doveri, conseguenze e ruoli

2. POLITICHE ETICHE PER VEICOLI AUTONOMI
L’etica delle conseguenze e il divieto di tortura/Priorità e soluzione di conflitti morali/Preferenze culturali e politiche etiche/Politiche libertarie e segregazioniste/Supererogazione, libertà di scelta e paternalismo morale/Guida autonoma e crisi climatica

3. AUTONOMIA DELL'ARTIFICIALE ED ETICA APPLICATA
Autoveicoli ad autonomia crescente/Autonomia operativa: dal termostato ai robot calciatori/Etica delle macchine autonome: scopi e strumenti/Etica delle macchine autonome: aspettative e problemi di fondo/Lo studio algoritmico dei giochi e le sue applicazioni inattese

4. METTERE AL BANDO LE ARMI AUTONOME?
Il ricercatore come vedetta etica/Dai droni alle armi autonome: problematiche etiche/Dai droni alle armi autonome: sviluppi tecnologici/Etica dei doveri: principi di distinzione e proporzionalità in guerra/Etica dei doveri: responsabilità e dignità dell’uomo/Etica delle conseguenze e armi autonome

5. POLITICHE ETICHE PER LE ARMI AUTONOME
Non tutto è permesso in guerra/La definizione di arma autonoma nel dibattito internazionale/Dalle definizioni al problema del controllo umano/La “condizione Petrov” nel contesto dell’intelligenza artificiale/Politiche etiche uniformi per il controllo umano/Politiche etiche differenziate e prudenziali

6. LA FORESTA DELLE MACCHINE AUTONOME E L'ETICA
Alan Turing sul futuro del lavoro/Norbert Wiener su automazione, etica e diritti dei lavoratori/Sorvegliare, educare e punire al tempo dell’intelligenza artificiale/Sorvegliare al tempo dei droni/Il chirurgo e il suo robot ad autonomia crescente/Scelte morali in cerca d’autore

This book is dedicated to expert research topics, and analyses of ethics-related inquiry, at the machine ethics and morality level: key players, benefits, problems, policies, and strategies. Gathering some of the leading voices that recognize and understand the complexities and intricacies of human-machine ethics early on in this phenomenon provides a resourceful compendium to be accessed by decision-makers and theorists concerned with identification and adoption of human-machine ethics initiatives, leading to needed policy adoption and reform for human-machine entities, their technologies, and their societal and legal obligations. The book encompasses theory and practice sections for each topical component of important areas of human-machine ethics both in existence today, and prospective for the future. Yet, human-machine ethics is not a futuristic matter; it is ever-present, here and now – thanks largely to ongoing technological advances in robotics, artificial intelligence, nanotechnology, and synthetic biology, among other allied fields.

In this paper, we analyse the structure of evolving moral dilemmas with an eye of regard for the increasing importance of the role of artificial intelligence in such context. Starting with the analysis of the famous trolley problem experiment as formulated by Philippa Foot, we consider subsequent variants of this moral dilemma conceived throughout the years, culminating with formulations of the trolley problem concerning artificial intelligence, in which self-driving vehicles will have to make life or death decisions autonomously. In doing so, we investigate the basis for the construction of dilemmatic questions both for humans and machines by considering the problem from a philosophical, social and neuroscientific perspective. After considering and analysing the trolley problem in utilitarian and deontological terms, we follow Rittel and Webber’s footsteps, by highlighting the fallacies of the deontological and utilitarian traditional ‘one-right-answer’ approach, where a solution i...

Given the possibility of a technological singularity  within our lifetimes, and the ‘black box’ nature of neural network outputs , is it not pertinent to investigate ways of subverting the possible implications for humanity should AI be deployed towards malevolent ends?

This essay aims to investigate the possible applicability of Ubuntu, an African moral theory which prioritises relationality in terms of moral recognition and responsibility towards one’s community , in relation to the benevolent deployment of AI for the benefit of all humanity. Further, this essay proposes that Ubuntu as a moral theory that succinctly embodies the original aims of the AI research community in that its primary concern in moral adjudication is the wellbeing of the community at large . Alternatively put, I hypothesize that Ubuntu offers an effective framework for maximising the potential beneficial application of AI and minimizing the potential for the malicious deployment of AI from within the AI research community.

From another perspective, more generally stated, this essay is a tentative investigation of the possibility of deploying Ubuntu in the development of moral machines and the implications thereof. Importantly, the focus of my essay cannot be on the moral machines themselves. This is due to the current impassibility of the syntactic-semantic divide – the challenge of programming semantic understanding into machines - which continues to plague research into the creation of artificial moral agents.

This chapter focuses on robotic deception and offers a general overview of the ethical debate on such issue. Deception is not only a complex human phenomenon that has interested philosophers, psychologists, political theorists, artists, scientists and other intellectuals throughout history, but also an opportunity widely exploited in many areas of technological design, such as in computer interface design and social robotics. After a brief theoretical introduction that accounts for the connection between technology and deception, deceptive design is explored first in the field of Human Computer Interaction (HCI)-more specifically, in captology (the science of persuasive technologies)-and secondly in Human Robot Interaction (HRI) and social robotics. Afterwards, the ethical admissibility of deception by design is put to test and the most relevant positions on the matter are reviewed with an eye to the philosophical background against which they are to be understood. It will be shown how deontological and consequentialist approaches are the most common methods of inquiry adopted in the debate and their respective strengths and weaknesses will be underlined. Finally, it will be discussed the ethical admissibility of consciously engaging in fictional relationships with robot companions and carers, which represents a relevant but also controversial case in this respect.

Much thought has been given to the question of who bears moral and legal responsibility for actions performed by robots. Some argue that responsibility could be attributed to a robot if it possessed human-like autonomy and metavolitionality, and that while such capacities can potentially be possessed by a robot with a single spatially compact body, they cannot be possessed by a spatially disjunct, decentralized collective such as a robotic swarm or network. However, advances in ubiquitous robotics and distributed computing open the door to a new form of robotic entity that possesses a unitary intelligence, despite the fact that its cognitive processes are not confined within a single spatially compact, persistent, identifiable body. Such a “nonlocalizable” robot may possess a body whose myriad components interact with one another at a distance and which is continuously transforming as components join and leave the body. Here we develop an ontology for classifying such robots on the basis of their autonomy, volitionality, and localizability. Using this ontology, we explore the extent to which nonlocalizable robots—including those possessing cognitive abilities that match or exceed those of human beings—can be considered moral and legal actors that are responsible for their own actions.

(note: this is a chapter in the book "Beyond Artificial
Intelligence: The Disappearing Human-Machine Divide," Eds. Jan Ramportl, Eva Zackova and Jozef Kelemen (Springer, 2014), pp.97-109).
ABSTRACT: The growing body of work in the new field of “affective robotics” involves both theoretical and practical ways to instill—or at least imitate—human emotion in Artificial Intelligence (AI), and also to induce emotions to-ward AI in humans. The aim of this is to guarantee that as AI becomes smarter and more powerful, it will remain tractable and attractive to us. Inducing emo-tions is important to this effort to create safer and more attractive AI because it is hoped that instantiation of emotions will eventually lead to robots that have moral and ethical codes, making them safer; and also that humans and AI will be able to develop mutual emotional attachments, facilitating the use of robots as human companions and helpers. This paper discusses some of the more sig-nificant of these recent efforts and addresses some important ethical questions that arise relative to these endeavors.

Road vehicle travel at a reasonable speed involves some risk, even when using computer-controlled driving with failure-free hardware and perfect sensing. A fully-automated vehicle must continuously decide how to allocate this risk without a human driver’s oversight. These are ethical decisions, particularly in instances where an automated vehicle cannot avoid crashing. In this chapter, I introduce the concept of moral behavior for an automated vehicle, argue the need for research in this area through responses to anticipated critiques, and discuss relevant applications from machine ethics and moral modeling research.

The field of "machine ethics" has raised the issue of the moral agency and responsibility of artificial entities, like computers and robots under the heading of "artificial moral agents" (AMA). In this article, we work through philosophical assumptions, conceptual and logical variations, and identify methodological as well as conceptual problems based on an analysis of all sides in this debate. A main conclusion is that many of these problems can be better handled by moving the discussion into a more outright normative ethical territory. Rather than locking the issue to be about AMA, a more fruitfull way forward is to address to what extent both machines and human beings should be in different ways included in different (sometimes shared) practices of ascribing moral agency and responsibility.

In this paper I argue that in certain circumstances robots can be seen as real moral agents. A distinction is made between persons and moral agents such that, it is not necessary for a robot to have personhood in order to be a moral agent. I detail three requirements for a robot to be seen as a moral agent. The first is achieved when the robot is significantly autonomous from any programmers or operators of the machine. The second is when one can analyze or explain the robot's behavior only by ascribing to it some predisposition or 'intention' to do good or harm. And finally, robot moral agency requires the robot to behave in a way that shows and understanding of responsibility to some other moral agent. Robots with all of these criteria will have moral rights as well as responsibilities regardless of their status as persons.

In scientific communications, journal articles, and philosophical aesthetic debates the words “art”, “creativity”, and “machine” are put together more and more frequently. Since some machines are designed to, or happens to, imitate human artistic creativity, it seems natural to use the same words to talk about human artists and machines which imitate them. However, the evolution of language in light of technology may conceal specific features of the phenomena it is supposed to describe. This makes it difficult to understand what machine creativity actually is and how it is connected to human creativity.
The aim of the paper is to investigate why, and in what sense, the functioning of some machines may be described in terms of artistic creativity and what is the relation between machine creativity and its human archetype. I start (§1) by introducing some general ideas concerning how language evolves alongside new technologies and focusing on the case of machine creativity. In §2 I review how some creative machines have been presented to the public, thus showing that a linguistic habit which connects machines to artistic creativity has already formed. In §3 I survey the debate on machine aesthetics and I highlight the main traits of the issue. In §4 I submit a primary scheme of machine creativity which draws on the concept of functional autonomy. Finally (§5), I argue that machine creativity cannot be mistaken for a substitute of the human kind, despite the fact that the word is the same. Creative machines are technological mediums by which new forms of human aesthetic experience can be explored. The kind of creativity machines enjoy displays a functional nature.

Artificial intelligence (AI) increasingly pervades all areas of life. To seize the opportunities this technology offers society, while limiting its risks and ensuring citizen protection, different stakeholders have presented guidelines for AI ethics. Nearly all of them consider similar values to be crucial and a minimum requirement for “ethically sound” AI applications – including privacy, reliability and transparency. However, how organisations that develop and deploy AI systems should implement these precepts remains unclear. This lack of specific and verifiable principles endangers the effectiveness and enforceability of ethics guidelines. To bridge this gap, this paper proposes a framework specifically designed to bring ethical principles into actionable practice when designing, implementing and evaluating AI systems.

List of Authors:
Dr Sebastian Hallensleben
Carla Hustedt
Lajla Fetic
Torsten Fleischer
Paul Grünke
Dr Thilo Hagendorff
Marc Hauer
Andreas Hauschke
PD Dr Jessica Heesen
Michael Herrmann
Prof. Dr Rafaela Hillerbrand
Prof. Emeritus Christoph Hubig
Dr Andreas Kaminski
Tobias Krafft
Dr Wulf Loh
Philipp Otto
Michael Puntschuh

Automated vehicles have received much attention recently, particularly the DARPA Urban Challenge vehicles, Google’s self-driving cars, and various others from auto manufacturers. These vehicles have the potential to significantly reduce crashes and improve roadway efficiency by automating the responsibilities of the driver. Still, automated vehicles are expected to crash occasionally, even when all sensors, vehicle control components, and algorithms function perfectly. If a human driver is unable to take control in time, a computer will be responsible for pre-crash behavior. Unlike other automated vehicles—such as aircraft, where every collision is catastrophic, and guided track systems, which can only avoid collisions in one dimension—automated roadway vehicles can predict various crash trajectory alternatives and select a path with the lowest damage or likelihood of collision. In some situations, the preferred path may be ambiguous. This study investigates automated vehicle crashing and concludes the following: (1) automated vehicles will almost certainly crash, (2) an automated vehicle’s decisions preceding certain crashes will have a moral component, and (3) there is no obvious way to effectively encode complex human morals in software. A three-phase approach to developing ethical crashing algorithms is presented, consisting of a rational approach, an artificial intelligence approach, and a natural language requirement. The phases are theoretical, and should be implemented as the technology becomes available.

Many researchers have argued that humanity will create artificial general intelligence (AGI) within the next twenty to one hundred years. It has been suggested that AGI may pose a catastrophic risk to humanity. After summarizing the arguments for why AGI may pose such a risk, we survey the field’s proposed responses to AGI risk. We consider societal proposals, proposals for external constraints on AGI behaviors, and proposals for creating AGIs that are safe due to their internal design.

Today, an increasing number and variety of autonomous robots and software applications are working in many areas of our daily life. The fear is that their complicated interactions with one another and with humans might lead to unpredictable disasters. Thus, some researchers insist that we should enable artificial systems to autonomously make a 'good' decision, i.e., a decision that respects society's values. Thus emerged a new academic field known as 'artificial morality,' 'artificial ethics', or 'machine ethics', a discipline that attempts to implement human moral judgement and moral action into artificial systems. According to some researchers in this field, it is sufficient to make an artificial agent only apparently moral, without regard for whether they are fully moral in the same sense as human beings. Just as weak artificial intelligence (AI) can be sufficient for certain practical purposes, limited 'functional morality' will be just as effective for certain practical purposes. Additionally, most projects concerning artificial morality are in line with mainstream artificial intelligence. In this paper, we critically consider whether this analogy between artificial intelligence and artificial morality holds true. We point out that one of the critical differences is that, while deploying artificial intelligence for certain practical purposes raises no question in general, to deploy artificial morality can be morally questionable in some situations.

Work in the field of machine medical ethics, especially as it applies to healthcare robots, generally focuses attention on controlling the decision making capabilities and actions of autonomous machines for the sake of respecting the rights of human beings. Absent from much of the current literature is a consideration of the other side of this issue. That is, the question of machine rights or the moral standing of these socially situated and interactive technologies. This paper investigates the moral situation of healthcare robots by examining how human beings should respond to these artificial entities that will increasingly come to care for us. A range of possible response s will be considered bounded by two opposing positions. We can, on the one hand, deal with these mechanisms by deploying the standard instrumental theory of technology, which renders care-giving robots nothing more than tools and therefore something we do not really need to care about. Or we can, on the other hand, treat these machines as domestic companions that occupy the place of another “person” in social relationships, becoming someone we increasingly need to care about. Unfortunately neither option is entirely satisfactory, and it is the objective of this paper not to argue for one or the other but to formulate the opportunities and challenges of ethics in the era of robotic caregivers.

Great technological advances in such areas as computer science, artificial intelligence, and robotics have brought the advent of artificially intelligent robots within our reach within the next century. Against this background, the interdisciplinary field of machine ethics is concerned with the vital issue of making robots "ethical" and examining the moral status of autonomous robots that are capable of moral reasoning and decision-making. The existence of such robots will deeply reshape our socio-political life. This paper focuses on whether such highly advanced yet artificially intelligent beings will deserve moral protection (in the form of being granted moral rights) once they become capable of moral reasoning and decision-making. I argue that we are obligated to grant them moral rights once they have become full ethical agents, i.e., subjects of morality. I present four related arguments in support of this claim and thereafter examine four main objections to the idea of ascribing moral rights to artificial intelligent robots.

We present a formalization and computational implementation of the second formulation of Kant's categorical imperative. This ethical principle requires an agent to never treat someone merely as a means but always also as an end. Here we interpret this principle in terms of how persons are causally affected by actions. We introduce Kantian causal agency models in which moral patients, actions, goals, and causal influence are represented, and we show how to formalize several readings of Kant's categorical imperative that correspond to Kant's concept of strict and wide duties towards oneself and others. Stricter versions handle cases where an action directly causally affects oneself or others, whereas the wide version maximizes the number of persons being treated as an end. We discuss limitations of our formalization by pointing to one of Kant's cases that the machinery cannot handle in a satisfying way.

Despite the rapid and continuous evolution of computer games, Articial Intelligence (AI) has not followed the same pace of development generating inappropriate behavior on virtual agents. There are numerous research fronts with approaches to reduce this problem, however, is dicult to get close of the expected behavior of human beings.

The position of the machine in contemporary culture is changing. At one time, these mechanisms were only objects to be used more or less effectively by human agents. Recent innovations in robotics, learning algorithms, and autonomous systems, however, challenge this standard position, presenting us with socially interactive machines that appear to be more than mere instruments. This essay examines this subtle but important shift in the social position and situation of machines. Toward this end, it is divided into four parts or movements: The first, examines the advantages and disadvantages of the instrumental view of technology—a theory that effectively situates the machine as a kind of intermediary through which human agents act on and in the world. The second part considers recent advancements in posthumanism and autonomous technology that challenge this tradition and provide good reasons to consider the machine otherwise. The third examines the opportunities and challenges of this transformation whereby what had been a mere technological object is now able to be recognized and dealt with as another kind of socially aware and interactive subject. And the fourth and final part, considers two ways that the conversation about machine moral standing might proceed from this point forward.