Marc Anderson

In this article we present a new approach to practical artificial intelligence (AI) ethics in heavy industry, which was developed in the context of an EU Horizons 2020 multi partner project. We begin with a review of the concept of Industry 4.0, discussing the limitations of the concept, and of iterative categorization of heavy industry generally, for a practical human centered ethical approach. We then proceed to an overview of actual and potential AI ethics approaches to heavy industry, suggesting that current approaches with their emphasis on broad high-level principles are not well suited to AI ethics for industry. From there we outline our own approach in two sections. The first suggests tailoring ethics to the time and space situation of the shop floor level worker from the ground up, including giving specific and evolving ethical recommendations. The second describes the ethicist’s role as an ethical supervisor immersed in the development process and interpreting between industrial and technological (tech) development partners. In presenting our approach we draw heavily on our own experiences in applying the method in the Use Cases of our project, as examples of what can be done.

AI has become the poster child for a certain kind of thinking which holds that some technologies can become objective, independent and emergent entities which can evolve beyond the control of their creators. This thinking is not new however. It is a product of certain philosophical ideas such as materialism, a common-sense world of objective and independent objects, a correspondence theory of truth, and so forth, which are centered around the pre-eminence of science, epistemology, and logical reasoning, among others, as the supremely valid modes of engaging experience. This paper aims to critically examine the synthesis of the development program for AI to be found in the writings of computer scientist John McCarthy (1927–2011). McCarthy has been called one of the founding fathers of AI. First, some of the main themes which recur in his writings from as early as the mid nineteen fifties in the Dartmouth proposal, up to his late writings, are considered. A discussion of the goals that such a program implies, follows, in terms of society’s relation with AI based technology and the nature and purpose of AI itself. These implications are arguably related to social control and are also moral. Ultimately, McCarthy’s idea that AI is and can be built upon a paradigm of reasoning and logic so as to simulate a perfectly rational abstraction of human thinking is shown to be motivated by his view that AI systems should be controlled, as servants. This understanding of human thinking is questionable, however, given current research on human irrationality, as well as on the argumentative role of reasoning. And yet, the process of development toward the unachievable goal of controllable, rational, logical AI systems, has handily come to serve as a blind for increasing techno-corporate control of society, with techno-corporate interests displacing blame for devaluations and harms caused by their push to develop the technology to AI systems which they can conveniently ‘lose control of.’

In this paper, we present the methodology we used in the European Horizon 2020 AI-PROFICIENT project, to evaluate the implementation of the ethical component of the project. The project is a 3-year collaboration between a university partner and industrial and tech partners, which aims to research the integration of AI services in heavy industry work settings. An AI ethics approach developed for the project has involved embedded ethical analysis of work contexts and design solutions and the generation of specific and evolving ethical recommendations for partners. We have performed an ongoing evaluation and monitoring of the implementation of recommendations. We describe the quantitative results of these implementations: overall, broken down by category, and broken down by category and responsible project partner (anonymized). In parallel, we discuss the results in light of our approach and offer insights for future research into the ground-level application of ethical recommendations for AI in heavy industry.

This article undertakes a comparative ethical analysis of the types of public expectations and concerns related to the development of two technologies: the telegraph and artificial intelligence. For each technology I provide a historical survey of public ethical expectations and concerns followed by a survey of the outcome or results of those expectations. Expectations and concerns of the telegraph era public are drawn together from popular and public literature and regulation of the period, whereas the expectations and concerns of our contemporary public AI engagement are drawn both from popular literature and public surveys, and supported by a manual search and ranking of a number of ethics related terms found in the raw feedback of the Stakeholder Consultation on the EU Commission High Level Expert Group Guidelines for Trustworthy AI. I then go on to compare those results, highlighting the similarities and differences between the two technologies, in particular the positive economic and socially responsible use expectations outcomes and the negative concerns regarding monopoly, regulation, and control. Finally, I argue that, taking the telegraph outcome as a guide, an ethical focus on accentuating positive expectations toward AI is more likely to produce definite results than concentrating upon prohibitory and negative approaches.

This paper explores the question of how we can know if Artificial Intelligence (AI) systems have become or are becoming sentient. After an overview of some arguments regarding AI sentience, it proceeds to an outline of the notion of negation in the philosophy of Josiah Royce, which is then applied to the arguments already presented. Royce’s notion of the primitive dyadic and symmetric negation relation is shown to bypass such arguments. The negation relation and its expansion into higher types of order are then considered with regard to how, in small variations of active negation, they would disclose sentience in AI systems. Finally, I argue that the much-hyped arguments and apocalyptic speculations regarding Artificial General Intelligence (AGI) takeover and similar scenarios, abetted by the notion of unlimited data, are based on a fundamental misunderstanding of how entities engage their experience. Namely, limitation, proceeding from the symmetric negation relation, expands outward into higher types of order in polyadic relations, wherein the entity self-limits and creatively moves toward uniqueness.