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

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

AI systems are bringing an augmentation of human capabilities to shape the world. They may also drag a replacement of human conscience in large chunks of life. AI systems can be designed to leave moral control in human hands, to obstruct or diminish that moral control, or even to prevent it, replacing human morality with pre-packaged or developed ‘solutions’ by the ‘intelligent’ machine itself. Artificial Intelligent systems (AIS) are increasingly being used in multiple applications and receiving more attention from the (...) public and private organisations. The purpose of this article is to offer a mapping of the technological architectures that support AIS, under the specific focus of the moral agency. Through a literature research and reflection process, the following areas are covered: a brief introduction and review of the literature on the topic of moral agency; an analysis using the BDI logic model (Bratman 1987); an elemental review of artificial ‘reasoning’ architectures in AIS; the influence of the data input and the data quality; AI systems’ positioning in decision support and decision making scenarios; and finally, some conclusions are offered about regarding the potential loss of moral control by humans due to AIS. This article contributes to the field of Ethics and Artificial Intelligence by providing a discussion for developers and researchers to understand how and under what circumstances the ‘human subject’ may, totally or partially, lose moral control and ownership over AI technologies. The topic is relevant because AIS often are not single machines but complex networks of machines that feed information and decisions into each other and to human operators. The detailed traceability of input-process-output at each node of the network is essential for it to remain within the field of moral agency. Moral agency is then at the basis of our system of legal responsibility, and social approval is unlikely to be obtained for entrusting important functions to complex systems under which no moral agency can be identified. (shrink)

The debate about the ethical implications of Artificial Intelligence dates from the 1960s :741–742, 1960; Wiener in Cybernetics: or control and communication in the animal and the machine, MIT Press, New York, 1961). However, in recent years symbolic AI has been complemented and sometimes replaced by Neural Networks and Machine Learning techniques. This has vastly increased its potential utility and impact on society, with the consequence that the ethical debate has gone mainstream. Such a debate has primarily focused on principles—the (...) ‘what’ of AI ethics —rather than on practices, the ‘how.’ Awareness of the potential issues is increasing at a fast rate, but the AI community’s ability to take action to mitigate the associated risks is still at its infancy. Our intention in presenting this research is to contribute to closing the gap between principles and practices by constructing a typology that may help practically-minded developers apply ethics at each stage of the Machine Learning development pipeline, and to signal to researchers where further work is needed. The focus is exclusively on Machine Learning, but it is hoped that the results of this research may be easily applicable to other branches of AI. The article outlines the research method for creating this typology, the initial findings, and provides a summary of future research needs. (shrink)

This paper provides a systematic literature review, analysis and discussion of methods that are proposed to practise ethics in research and innovation. Ethical considerations concerning the impacts of R&I are increasingly important, due to the quickening pace of technological innovation and the ubiquitous use of the outcomes of R&I processes in society. For this reason, several methods for practising ethics have been developed in different fields of R&I. The paper first of all presents a systematic search of academic sources that (...) present and discuss such methods. Secondly, it provides a categorisation of these methods according to three main kinds: ex ante methods, dealing with emerging technologies, intra methods, dealing with technology design, and ex post methods, dealing with ethical analysis of existing technologies. Thirdly, it discusses the methods by considering problems in the way they deal with the uncertainty of technological change, ethical technology design, the identification, analysis and resolving of ethical impacts of technologies and stakeholder participation. The results and discussion of our literature review are valuable for gaining an overview of the state of the art and serve as an outline of a future research agenda of methods for practising ethics in R&I. (shrink)

This article analyzes recent cases of company-sponsored online experiments with unsuspecting users and discusses the ethical aspects of such experimentation. These cases illustrate a new type of online research where companies modify their algorithms to intentionally misinform or mislead users. Unlike typical forms of A/B testing, where two versions of the same website are presented to different users to evaluate interface changes, algorithm modification is a deeper form of testing where changes in program code induce user deception. Thus, we propose (...) to call this new approach C/D experimentation to distinguish it from the surface-level website evaluation associated with A/B testing. Three aspects raise ethical concerns regarding C/D experimentation: the absence of user consent to participate in research, the presence of intentional deception, and the complete lack of protection for human subjects who partake in privately funded behavioral research. Three recommendations are proposed to address these issues: to develop an ethical code of conduct for subject protection shared by online companies, to include special provisions for C/D experiments in social networking platforms, and to create an independent user advocacy board to protect the rights of users who partake in online research conducted in the private sector. (shrink)

This paper aims at investigating the perceived difference between ethics and IT ethics in college students. The study mainly investigates whether university students in the Middle East and their counterpart in the USA hold the same ethical values both in a traditional context and in an IT context. The study also investigates possible differences in students’ ethics considering their level of study and whether they have prior business ethics knowledge or not. Furthermore, the study controls for possible self-others bias in (...) students’ responses to ethical issues by addressing some ethical issues where the respondent is the subject of the issue and other ethical issues where someone other than the respondent is the subject of the issue. Questionnaires were administered to university students in the USA and in several Middle East countries. A total of 401 usable questionnaires were returned. The study found that both the USA and the Middle East students have statistically significant lower ethical intention scores when addressing ethical issues in an IT context compared to traditional non-IT context. This pattern was consistent regardless of the students’ level of study and their prior knowledge of business ethics, and taking into consideration self-others bias. Additionally, it was found that graduate students have higher IT as well as non-IT ethical intentions compared to undergraduate students, however, it was found that the students’ prior business ethics knowledge did not affect the students’ ethical intention scores. Self-other bias was found to act in opposite directions for ethical issues in an IT context compared to ethical issues in a traditional context. For ethical issues within IT-context, students were less strict with themselves while they were stricter with others. For ethical issues within traditional non-IT context, students were stricter with themselves, while they were less strict with others. Finally, the study found that while students both in the USA and in the Middle East disagree with unethical IT as well as non-IT conducts, the degree to which they disagree with the unethical behavior significantly differ, where USA students were found to be significantly stricter with their disagreement. (shrink)

Attempts to employ discourse ethics for assessing communication and information technologies have tended to focus on managerial and policy-oriented contexts. These initiatives presuppose institutional resources for organizing sophisticated consultation processes that elicit stakeholder input. Drawing on Jürgen Habermas’s discourse ethics, this paper supplements those initiatives by developing a more widely usable framework for moral inquiry and reflection on problematic cyberpractices. Given the highly idealized character of discourse ethics, a usable framework must answer two questions: How should those who lack organizational (...) power conduct their moral inquiry under non-ideal conditions of discourse? How ought they to understand the moral force of the judgments they reach under such conditions? In response, I develop the heuristic implications of Habermas’s principle of universalization. To render that principle usable for non-ideal discourse, I propose a modification that yields a scalar measure of “dialogically robust” judgments that are responsive to the actual state of discussion. To illustrate the use of these principles, I sketch two case studies, which examine the moral acceptability of violent video gaming and government cyber-surveillance. (shrink)

Data science, and the related field of big data, is an emerging discipline involving the analysis of data to solve problems and develop insights. This rapidly growing domain promises many benefits to both consumers and businesses. However, the use of big data analytics can also introduce many ethical concerns, stemming from, for example, the possible loss of privacy or the harming of a sub-category of the population via a classification algorithm. To help address these potential ethical challenges, this paper maps (...) and describes the main ethical themes that were identified via systematic literature review. It then identifies a possible structure to integrate these themes within a data science project, thus helping to provide some structure in the on-going debate with respect to the possible ethical situations that can arise when using data science analytics. (shrink)

Biometric technology was once the purview of security, with face recognition and fingerprint scans used for identification and law enforcement. This is no longer the case; biometrics is increasingly used for commercial and civil applications. Due to the widespread diffusion of biometrics, it is important to address the ethical issues inherent to the development and deployment of the technology. This article explores the burgeoning research on biometrics for non-security purposes and the ethical implications for organizations. This will be achieved by (...) reviewing the literature on biometrics and business ethics and drawing from disciplines such as computer ethics to inform a more robust discussion of key themes. Although there are many ethical concerns, privacy is the key issue, with associated themes. These include definitions of privacy, the privacy paradox, informed consent, regulatory frameworks and guidelines, and discrimination. Despite the proliferation of biometric technology, there is little empirical research on applied biometrics and business ethics. As such, there are several avenues for research to improve understanding of the ethical implications of using this technology. (shrink)

Empirical research into the ethics of emerging technologies, often involving foresight studies, technology assessment or application of the precautionary principle, raises significant epistemological challenges by failing to explain the relative epistemic status of contentious normative claims about future states. This weakness means that it is unclear why the conclusions reached by these approaches should be considered valid, for example in anticipatory ethical assessment or governance of emerging technologies. This paper explains and responds to this problem by proposing an account of (...) how the epistemic status of uncertain normative claims can be established in ethical and political discourses based on Jürgen Habermas’ discourse ethics. To better understand the nature of the problem, the relationship between norms, facts and the future is explored in light of potential meta-ethical fallacies faced in the field of empirical ethics. Weaknesses of current approaches to anticipatory ethical assessment and governance are then explored, including the Precautionary Principle and Technology Assessment. We argue that the epistemic status of uncertain normative claims can be understood within Habermas’ approach to political discourse, which requires ‘translation’ of uncertain claims to be comprehensible to other stakeholders in discourse. Translation thus provides a way to allow for uncertain normative claims to be considered alongside other types of validity claims in discourse. The paper contributes a conceptual account of the epistemic status of uncertain normative claims in discourse and begins to develop a ‘methodology of translation’ which can be further developed for approaches to research and ethical assessment supporting anticipatory evidence-based policy, governance and system design. (shrink)

Information security can be of high moral value. It can equally be used for immoral purposes and have undesirable consequences. In this paper we suggest that critical theory can facilitate a better understanding of possible ethical issues and can provide support when finding ways of addressing them. The paper argues that critical theory has intrinsic links to ethics and that it is possible to identify concepts frequently used in critical theory to pinpoint ethical concerns. Using the example of UK electronic (...) medical records the paper demonstrates that a critical lens can highlight issues that traditional ethical theories tend to overlook. These are often linked to collective issues such as social and organisational structures, which philosophical ethics with its typical focus on the individual does not tend to emphasise. The paper suggests that this insight can help in developing ways of researching and innovating responsibly in the area of information security. (shrink)

Purpose– The purpose of this paper is to present a novel mnemonic, ACTIVE, inspired by Mason's 1985 PAPA mnemonic, which will help researchers and IT professionals develop an understanding of the major issues in information ethics.Design/methodology/approach– Theoretical foundations are developed for each element of the mnemonic by reference to philosophical definitions of the terms used and to virtue ethics, particularly MacIntyrean virtue ethics. The paper starts with a critique of the elements of the PAPA mnemonic and then proceeds to develop (...) an understanding of each of the elements of ACTIVE ethics, via a discussion of the underpinning virtue ethics.Findings– This paper identifies six issues, described by the mnemonic, ACTIVE. ACTIVE stands for: autonomy, the ability of the individual to manage their own information and make choice; community, the ethical effect of an information systems on the community which it supports; transparency, the extent to which the derivation of content and process in an information system is made clear; identity, the social and ethical effect of an information system on the definition and maintenance of the distinctive characteristics of a person; value, the value or moral worth placed on information associated with an individual and hence on the relationship with the individual; and empathy, the ability of the information systems professional to emotionally connect with the user and the extent to which the information system distances or connects.Originality/value– The paper applies virtue ethics to developing a tool to help information professionals reflect on their ethical practice in developing and supporting information systems. (shrink)

Until recently, businesses and corporations could argue that their only real commitments were to maximise the return to their shareholders whilst staying within the law. However, the world has changed significantly during the last ten years and now most major corporations recognise that they have significant responsibility to local and global societies beyond simply making profit. This means that there is now an increasing concern with the question of how corporations, and their employees, ought to behave, and this leads us (...) to consider ethics as the appropriate theoretical and philosophical domain.I will bring into the debate two relatively recent approaches to ethics: Jürgen Habermas's discourse ethics ; and the critical realist approach of Roy Bhaskar. These are interesting for several reasons: they both draw on traditional ethical theories, although different ones; they bring in innovations of practical relevance; and they both share an over-arching critical perspective. After a critical introduction to both ethical theories, their similarities and differences are explored. The article ends by considering the extent to which they may be practically useful within business. (shrink)