Challenges of Explaining the Behavior of Black-Box AI Systems

摘要

There are many examples of problems resulting from inscrutable AI systems, so thereis a growing need to be able to explain how such systems produce their outputs. Drawingon a case study at the Danish Business Authority, we provide a framework andrecommendations for addressing the many challenges of explaining the behavior ofblack-box AI systems. Our findings will enable organizations to successfully developand deploy AI systems without causing legal or ethical problems.