Extrapolating Modeling Techniques for Machine Ethics Reasoning from AI & Law

Abstract

With the rise of Artificial Intelligence in a large variety of disciplines in science and industry the problem of ethical AI has become increasingly prevalent. As artificially intelligent agents grow in independence it becomes increasingly important to have guarantees about their behavior, to prevent them from acting unethically without needlessly prohibiting ethical behavior. To this end it is necessary for a formal model of reasoning with norms and values to be developed. One way of approaching this challenge is from the field of AI & law. After all, laws and legal procedures are already a type of encoding of ethics, and as such reasoning about and with them has clear parallels with formal models of ethics. This thesis aims to bridge the gap between the legal and the ethical to exploit and adapt the older and more refined field of AI & law to aid and augment decision systems from machine ethics.