From GPUs to GDP: How Data Center Economics Shapes AI Diffusion

4:30–5:00 Networking and refreshments 5:00–6:00 Lecture and Q&A session Most of us encounter and interact with different pieces of the Artificial Intelligence (AI) supply chain. Engineers make Graphics Processing Units (GPUs) and coders deploy Transformers. Utilities supply megawatts. County commissioners approve land, water, and tax abatements. Investors talk about capital expenditure. Managers track productivity from coding assistants. Customers talk to chatbots. Somewhere in between all of them sit enormous, expensive, mostly windowless buildings. Those data centers are the less visible part of the AI supply chain. Understanding their economic determinants helps explain many of this supply chain's extraordinary features, such as the timing of its rise, the bottlenecks holding back investment, and the risks associated with the current build-out. Enormous investment decisions are being made now, while the economic payoff will arrive years from now. Shane Greenstein, Ph.D., is the Martin Marshall Professor of Business Administration. He teaches in the Technology, Operations, and Management Unit. Encompassing a wide array of questions about microelectronics, computing, communication, and internet markets, Professor Greenstein's research extends from economic measurement and analysis to broader issues. His most recent book, How the Internet Became Commercial (2015, Princeton University Press), won the 2016 Schumpeter Prize for best book. Many media outlets cover his work. Professor Greenstein previously taught at the Kellogg School of Management at Northwestern University and at the University of Illinois, Urbana-Champaign. He received his Ph.D. from Stanford University and his BA from the University of California at Berkeley, both in economics. He continues to receive daily life lessons from his wife and children.