Artificial Intelligence, Employment and Income

Artificial intelligence (AI) will have many profound societal effects It promises potential benefits (and may also pose risks) in education, defense, business, law, and science In this article we explore how AI is likely to affect employment and the distribution of income. I am grateful for the helpful comments provided by many people Specifically I would like to acknowledge the advice teceived from Sandra Cook and Victor Walling of SRI, Wassily Leontief and Faye Duchin of the New York University Institute for Economic Analysis, Margaret Boden of The University of Sussex, Henry Levin and Charles Holloway of Stanford University, James Albus of the National Bureau of Standards, and Peter Hart of Syntelligence Herbert Simon, of Carnegie-Mellon Univetsity, wrote me extensive criticisms and rebuttals of my arguments Robert Solow of MIT was quite skeptical of my premises, but conceded nevertheless that my conclusions could possibly follow from them if certain other economic conditions were satisfied. There are two opposing views in response to this question Some claim that AI is not really very different from other technologies that have supported automation and increased productivity-technologies such as mechanical engineering, ele&onics, control engineering, and operations rcsearch. Like them, AI may also lead ultimately to an expanding economy with a concomitant expansion of employment opportunities. At worst, according to this view, thcrc will be some, perhaps even substantial shifts in the types of jobs, but certainly no overall reduction in the total number of jobs.