Macro Ethics for Governing Equitable Sociotechnical Systems

The evolving relationship between humans and technology entails increasing concerns about the impact on ethical issues such as bias, unfairness, and lack of accountability. There is thus a need for consistent responses to multiple-user social dilemmas that arise during interactions in sociotechnical systems, where combinations of humans and technical agents work together as ethical duos. The outcomes of these systems can be evaluated by the values that participating humans hold, which in turn influence the development of norms used to guide acceptable behaviour. However, when values are misjudged or norms conflict, dilemmas arise that must be resolved in satisfactory ways. To examine these dilemmas, we adopt a macro ethics perspective where ethics is addressed via the governance of sociotechnical systems with multiple agents (rather than through the actions of single agents). We propose that to produce satisfactory outcomes, systematic methodologies be developed to consistently integrate normative ethical principles in reasoning capacities. The application of these ethical principles would enable practitioners to think analytically and systematically about the multiple-user social dilemmas that occur in these systems, in order to resolve them in satisfactory ways. To achieve this, we need to (1) categorize ethical principles not yet used in AI, form new ways of (2) systematically integrating ethical principles into reasoning, and use these new ways to (3) develop consistent responses to multiple-user social dilemmas.