CMU Researchers Develop 'Electrotech Moneyball' Framework To Rank Risk and Opportunity in Energy and AI Supply Chains

Carnegie Mellon University researchers developed a framework(opens in new window) to help the United States secure the power infrastructure needed to support artificial intelligence and other emerging technologies, all while reducing vulnerabilities in the global supply chains that make that infrastructure possible.

The research, conducted over the past two years by the Carnegie Mellon Institute for Strategy and Technology (CMIST)(opens in new window), examines how competition among global powers is increasingly tied to two foundational resources: electricity and compute. As the United States expands its power grid and builds the data centers needed to support AI, the hardware and software connecting those systems will play an increasingly important role in national security, the researchers said.

Harry Krejsa(opens in new window), director of studies at CMIST, described the massive investment in new energy infrastructure as a “once-in-a-generation” opportunity to build security into the grid from the start rather than trying to address vulnerabilities later.

“The technologies redefining the battlefield and economic competitiveness alike increasingly draw on a common industrial base of batteries, power electronics, electric motors and embedded software,” Krejsa said.

Krejsa authored the paper with CMIST nonresident fellows Phoebe Benich(opens in new window) and Emma Stewart(opens in new window) in partnership with the Foundation for Defense of Democracies. The researchers draw on the same risk-prioritization logic used in defense manufacturing, where producers regularly separate the "smart," susceptible components that demand scrutiny from “dumb” commodity inputs.

Not every component needs the same level of security, the authors argue. The U.S. cannot stop using all Chinese parts right away, as blocking everything would slow energy expansion and harm power reliability. The “Moneyball” framework outlined in the paper assesses each technology across three dimensions to maximize strategic return: urgency of deployment, level of risk versus opportunity to mitigate and competitive advantage for domestic leadership.

For example, basic parts like batteries and solar panels don't pose the same security risks as the 'smart' software that runs them, such as the systems that manage energy storage, power inverters and cloud connections. These digital systems are what actually control how power is generated, stored, and sent across the grid.

Based on that assessment, the framework sorts components into three tiers of policy priority: tight domestic control, trusted-ally sourcing, or managed global procurement. 

The framework has circulated among the executive branch, Congress, allied ambassadors and industry, generating requests for briefings and testimony as well as interest from tech companies exploring how the approach could be used in their own sectors.

Looking at the big picture, Krejsa highlighted the central goal of this work.

“The nations that can build, power and secure this ‘electrotech’ foundation will hold the advantage across all of the technologies built on top of it, even as those same technologies put unprecedented capability into more hands than ever before — a paradox of concentrating foundations and diffusing power that CMIST is making a defining focus of its research,” he said.