Washington, Big Tech Clash Over AI Data Centers is an International Security Issue

Mica Udani, Ph.D., ACWPS Fellow

For U.S. policy makers, AI is an international security issue as well as a national security and domestic regulation challenge. The AI boom has shaken global economies, from energy infrastructure to labor and employment. In an effort to meet AI demand, Big Tech may be in a “race to the bottom” in global data center investment. By seeking cheaper, less regulated sites overseas, U.S. tech companies may exacerbate existing energy and water vulnerabilities in partner nations where they intend to host new data centers. 

Demand for generative AI outpaces data center capacity. While the technology industry races to build more data centers to meet demand, national governments are weighing various energy policies to keep up. This has put multinational corporations (MNCs) and national governments at a crossroads in the scramble to profit from the AI gold rush. 

For the U.S., it means the goal of maintaining data center dominance is threatened by Big Tech’s investing in data centers overseas. Synergy Research Group reports that roughly 60 percent of the world’s 1,244 largest data centers were outside the U.S. as of June 2025, and that at least 575 more data center projects are in development globally. Furthermore, India and Brazil–countries known for struggling with recurrent water shortages and energy insecurities–provide cheaper options for Big Tech and avoid some American regulatory controls. In other words, going overseas doesn’t just jeopardize American AI dominance, a national security priority, it will set back partner nation stability, a U.S. international security priority.

National security issues: economic and tech dominance 

Slow by comparison to the demand and investment in AI is the response from research and advocacy groups to raise questions about the ill effects for the global economy and the people who power it. Among the problems raised so far include data center electricity consumption, water consumption, and energy insecurity and infrastructure. Data center capacity is expected to triple by 2030. The Lawrence Berkeley National Laboratory reports that electricity generation in the U.S. must expand significantly to meet the surging electricity demand from AI use, even under the more conservative estimates before the AI boom in 2025. Data centers consumed about 4.4 percent of total U.S. electricity demand in 2023, but by 2028, will consume 7 percent to 12 percent of the nation’s electricity demand. 

In the U.S., the negative effects of data centers are already hotly protested, especially in communities in the Northern Virginia “Data Center Alley” where they are currently concentrated. Major issues raised by citizens include overheating and nonstop noise of data centers; overheating requires cooling solutions, all of which consume a vast amount of energy and water.  Studies demonstrate that electricity prices across the country found that monthly electricity costs have gone up as much as 267% compared with five years ago in locations near substantial data center activity, taking a toll on families in these communities. At the state level, policymakers in areas where data centers are being built, from New Jersey to Wisconsin, are responding to citizen concerns by considering greater regulatory legislation on data centers and improving clean energy supplies, in an effort to offset the impacts of data centers on local communities.

Some regulatory controls push tech companies overseas, however. The Trump administration’s recent executive orders are meant to decrease unnecessary regulations. Since they focus on federal land development, they sidestep issues associated with data centers in populated residential areas. The administration aims to expand the development of data center infrastructure in the U.S., as well as to promote U.S. dominance of the global AI market. These initiatives are meant to speed up environmental regulatory reviews and speed development on federal land. Despite these initiatives, technology firms are increasingly looking outside the U.S. for data center development opportunities.

International security issues: partner nation economic and political stability 

Major American technology MNCs Google, Meta, and Microsoft, among many others are increasingly outsourcing data center development abroad to take advantage of foreign direct investment (FDI) deals with other governments for cheap land, tax incentives, and lax regulations. Google announced a 15 billion USD investment last month for building data centers in India, following existing multi-billion dollar investments in Indian data centers from Amazon and Microsoft, despite increasing U.S.-India trade tensions. Technology MNCs are courting Brazil, India, Kenya, Singapore, and the UAE for further data center development investment as these states offer favorable FDI incentives.

However, the destination choice of these overseas data centers may have immense negative consequences on local communities, as many of these countries are known for energy and water insecurity and fragile infrastructure. Brazil and India in particular are vulnerable environments which already frequently suffer from water crises, drought, frequent blackouts, and electricity shortages. The huge energy and water consumption required by data centers may further disrupt already insecure infrastructure in these countries.  

Existing cases demonstrate how centers exacerbate instability in water and energy infrastructure. Amazon and Microsoft’s data centers in Mexico are in drought-prone regions with frequent blackouts. Residents near Microsoft’s Mexican data centers describe increasingly frequent power cuts and lengthier water outages, which have severely damaged public health. 

Extensive research on water and electricity crises demonstrates severe negative effects specific to women and children. Already, reports from villages near Microsoft’s data centers in Mexico report increased public health strain, from outbreaks due to constrained water access and local doctors treating children by flashlight. The negative effects on maternal and child health and family well being means the centers could have generational negative effects on security and stability. 

Way ahead

American policymakers and U.S. tech firms are at cross purposes. Policymakers want to maintain U.S. data center dominance and Big Tech is looking to take advantage of foreign government incentives of cheaper AI development. To maintain national and international security at a precarious time of great power competition and widespread societal instability and war, however, stakeholders must get this right. Going forward, decision  makers must ensure safe AI development that improves local economies without weakening energy and water infrastructure. Though MNCs may see better short-term profit with foreign incentives for data center development, they cannot risk investing in country contexts with fragile infrastructure and vulnerable populations.

To avoid an international race to the bottom in the negative effects of data centers on local infrastructure, MNCs and U.S. policymakers must commit to finding a measured, centralized approach:

• Local and State governments in the U.S. should provide data on impacts of data centers on regional populations and economies.

• Washington should help fast track research on the long-term effects of data centers in populated areas.

• Local, State, and national governments should strengthen clean energy infrastructure in regions selected for data center development. 

American MNCs must come to terms with the international security impacts of their bottom line. Investing in a long term risk mitigation strategy can help defray long-term insecurity of investing in data center development in vulnerable regions abroad. 

Catalina “Mica” Udani is a fellow at the American Council on Women Peace & Security and Senior Affiliate for the Partnership for Innovation, Cross-Sector Collaboration, Leadership, and Organization (PICCLO) at the University of Pennsylvania. She has a PhD in Political Science from the University of Pennsylvania and researches international relations, human rights, political economy, and migration.