Being Considerate as a Pathway to Pluralist Alignment

Pluralistic alignment is concerned with ensuring that an AI system’s objectives and behaviors are in harmony with the diversity of human values and perspectives. In this paper we study the notion of pluralistic alignment in the context of agentic AI, and in particular in the context of an agent that is trying to learn a policy in a manner that is mindful of the values and perspective of human agents it considers to be stakeholders of its environment. To this end, we show how being considerate of the future wellbeing and agency of other (human) agents can promote a form of pluralistic alignment.