AI Compute Demand Must Reach $6T by 2031 for Colossal Data Center Buildout

According to a recent analysis by Bain & Company, the demand for AI compute infrastructure must reach $6 trillion in annual revenue by 2031 to justify the massive data center buildout. This estimate is based on the expectation that hyperscalers such as Microsoft MSFT, Google GOOGGOOGL, Amazon AMZN, Meta META, and Oracle ORCL will continue to spend heavily on new infrastructure.

The report states that capital expenditures from these companies are expected to reach $780 billion this year, a five-fold increase from last year. The scale of these data centers is growing exponentially, with leading AI data centers approaching 1 GW capacity, and the world's largest campuses expected to reach 9 GW by 2030.

To justify such massive investment, Bain estimates that the AI market must generate $6 trillion in revenue annually by 2031. This will require innovation in new use cases, including consumer and enterprise AI markets, as well as emerging fields such as autonomous everything, physical AI, and new product development.