Cultural Representativeness in the Principles of AI

Artificial intelligence (AI) applications have reached a wide range of domains and have raised concerns over fairness, accountability, transparency, and ethics. For example, social media platforms face challenges from Congress over data privacy and facial recognition software have been racially biased. Accordingly, society is actively establishing principles to govern the development and application of AI technologies; examples include General Data Protection Regulation (GDPR). But, as AI innovation disseminates across cultural and political boundaries, how do societies in different cultures perceive these high-level AI principles? What are the acceptable ground rules for global AI governance? This project seeks to answer these by studying the interactions between cultural norms, the public opinion of AI, and the AI research community. To understand how key AI principles resonate amongst different cultures, we will study knowledge production and dissemination. Our approach is informed by the studies in comparative philosophy, which contrast moral traditions developed along relatively isolated cultural and regional lines. We will combine data science, qualitative studies, and mixed-methods approach to analyze micro- and macroscopic data. This project will build synergy among distinct disciplines represented by the team, including Philosophy and Ethics, Data and Information Sciences, Psychology, and Anthropology