Computing devices have become indispensable to modern life, yet they remain largely ignorant about the people they serve and the world they have so deeply permeated. Our goal is give computers "common sense"�the ability to understand and think about the everyday world in ways that people do. Our work is focused on building computational models of human commonsense psychology, especially models of how people make inferences about ordinary situations, activities and intentions. To do this we are developing a suite of commonsense knowledge bases, inference engines, and architectural elements for linking these together, as well as new kinds of applications built on these technologies. By giving machines such common sense, they will be able to better help us understand the world and ourselves, help us relate to machines and to each other, help us cope with our many problems, and help us become who we want to become both as individuals and as a society.