This article is about those people who are currently developing AI, less about the AI itself. There needs to be a paradigm shift in the philosophical assumptions that those who develop AI are committed to. I give a history of where those assumptions come from, and what their limiting consequences are.