By Doug Kelly and Kent Kaiser
If the United States is serious about leading the world in artificial intelligence, it must confront a difficult truth: AI is not just about algorithms, venture funding, cloud platforms, or research labs. It is about energy and power grids, natural resources and supply chains, and human capital.
Right now, the U.S. is unprepared for what is coming. China is rapidly integrating energy expansion, data center capacity, human capital, and global AI exports into a coherent national strategy.
The U.S. currently lacks the dependable, affordable, and scalable power needed for sustained AI development, a recent study by the American Edge Project showed. China, by contrast, has aggressively invested in energy and transmission for years, and as a result, is producing twice as much electricity as the U.S.
Contrary to common belief, the data centers that power AI aren’t the cause of the energy challenge we face — they merely expose three decades of underinvestment in America’s transmission grid.
Closing that gap requires an all-of-the-above approach to power generation — nuclear, natural gas, renewables, and advanced energy sources. We must onshore critical supply chains by incentivizing domestic production of gas turbines and power transformers to eliminate multiyear wait times. And we must modernize and secure the grid by streamlining permitting, expanding transmission, and protecting infrastructure against cyber threats. Utilities, regulators, developers, and communities must come together to fast-track the power projects critical to America’s AI future.
Supply chains are equally vulnerable. Every layer of AI infrastructure, from chips to networking switches to heat exchangers, depends on global supply chains that the U.S. does not fully control. More than 90% of the world’s most advanced semiconductors are produced in Taiwan, leaving us exposed to supply disruptions, political upheaval, or economic coercion. Rare earth processing remains overwhelmingly concentrated in China. JP Morgan’s recent $1.5 trillion commitment toward critical minerals underscores how seriously markets are taking these vulnerabilities — and so must we.
Continued investment in semiconductor manufacturing and AI accelerator development is essential to maintaining the U.S.’s compute advantage. We must secure critical mineral supply chains by developing domestic rare earth processing and diversifying sources beyond China’s chokehold. And we must prevent harmful regulation by enacting a multiyear freeze on state AI laws, rejecting restrictive copyright theories, and resisting antitrust experiments that would weaken U.S. innovators.
Continue reading at the Washington Examiner