Autonomous Agents and Ethical Decision-Making

Machine ethics, also known as artificial morality, is a newly emerging field concerned with ensuring appropriate behavior of machines toward humans and other machines. In this article, we discuss the importance of machine ethics and present a computational model of ethical decision-making for autonomous agents. The proposed model implements a mechanism for integrating the results of diverse assessments into a unique cue, and takes into account the agent’s preferences, good and bad past experiences, ethical rules, and current emotional state as the main factors involved in choosing the most appropriate option. The design of the model is based on theories and models developed in fields such as neuroscience, psychology, artificial intelligence, and cognitive informatics. In particular, the model attempts to emulate neural mechanisms of the human brain involved in ethical decision-making.