Gonzalo Génova, Valentín Moreno, and M. Rosario González explore the possibility and limitations of teaching ethical behavior to artificial intelligence. The paper delves into two main approaches to teaching ethics to machines: explicit ethical programming and learning by imitation. It highlights the difficulties faced by each approach and discusses the implications and potential issues surrounding the application of machine learning to ethical issues.
The authors begin by examining explicit ethical programming, such as Asimov’s Three Laws, and discuss the challenges involved in foreseeing the consequences of an act, as well as the necessity of having an explicit goal for ethical behavior. The second approach, learning by imitation, involves machines observing the behavior of experts or a majority in order to emulate them. The paper also discusses the Moral Machine experiment by MIT, which aimed to teach machines to make moral decisions based on the preferences of the majority.
Despite the potential of machine learning techniques, the authors argue that both approaches fail to capture the essence of genuine ethical thinking in human beings. They emphasize that ethics is not about following a code of conduct or imitating the behavior of others, but rather about critical thinking and the formation of one’s own conscience. The paper concludes by questioning whether machines can truly learn ethics like humans do, suggesting that current methods of teaching ethics to machines are inadequate for capturing the complexity of human ethical life.
The research presented in the paper raises important philosophical questions about the nature of ethics and the role of machines in our ethical lives. It challenges the instrumentalist and reductionist approaches to ethics, which view ethical values as computable or reducible to a set of rules. By highlighting the limitations of these approaches, the paper invites us to reconsider the importance of value rationality and the recognition of the uniqueness and unrepeatable nature of human beings in ethical considerations.
In light of these findings, future research could explore alternative approaches to teaching ethics to machines that go beyond mere rule-following or imitation. This could involve the development of novel machine learning techniques that foster critical thinking and the ability to reason with values without reducing them to numbers. Additionally, interdisciplinary collaboration between philosophers, AI researchers, and ethicists could further enrich our understanding of the ethical dimensions of artificial intelligence and help to develop AI systems that not only do the right thing but also respect the complexity and richness of human ethical life.
Abstract
Is ethics a computable function? Can machines learn ethics like humans do? If teaching consists in no more than programming, training, indoctrinating… and if ethics is merely following a code of conduct, then yes, we can teach ethics to algorithmic machines. But if ethics is not merely about following a code of conduct or about imitating the behavior of others, then an approach based on computing outcomes, and on the reduction of ethics to the compilation and application of a set of rules, either a priori or learned, misses the point. Our intention is not to solve the technical problem of machine ethics, but to learn something about human ethics, and its rationality, by reflecting on the ethics that can and should be implemented in machines. Any machine ethics implementation will have to face a number of fundamental or conceptual problems, which in the end refer to philosophical questions, such as: what is a human being (or more generally, what is a worthy being); what is human intentional acting; and how are intentional actions and their consequences morally evaluated. We are convinced that a proper understanding of ethical issues in AI can teach us something valuable about ourselves, and what it means to lead a free and responsible ethical life, that is, being good people beyond merely “following a moral code”. In the end we believe that rationality must be seen to involve more than just computing, and that value rationality is beyond numbers. Such an understanding is a required step to recovering a renewed rationality of ethics, one that is urgently needed in our highly technified society.
Machine Ethics: Do Androids Dream of Being Good People?