Bloomberg Technology

US Needs to Invest More in AI Buildout, Says Cato’s Frazier

The interview with Cato Institute’s Frazier argues that the United States must treat AI infrastructure as a national challenge, comparable to highways or ports, and accelerate the build‑out of data centers to support health care, commerce and defense. He highlights bottlenecks in permitting across 11 states, the dominance of China in chips and other components, and the scarcity of helium from Qatar, calling for a coordinated supply‑chain strategy with allies and substantial federal funding. Frazier dismisses myths such as data‑center heat‑island effects, citing Andy Masley’s research, and points to examples like Oklahoma’s “behind‑the‑meter” renewable power and Gavin McCormick’s work on grid‑resilient AI training to illustrate practical solutions. Without decisive investment, the U.S. risks falling behind in AI‑driven national security and economic competitiveness, while consumers could face higher electricity costs; policymakers are urged to prioritize resilient, clean AI infrastructure now.