The rapid expansion of AI infrastructure is beginning to affect American universities physically, as valuable university-owned land becomes attractive to companies building data centers. Fortune reports that George Washington University sold its roughly 120-acre Virginia Science and Technology Campus in Ashburn to Amazon Data Services for $427 million. The site is located in Northern Virginia’s “Data Center Alley,” where access to electricity, water, and fiber infrastructure makes land particularly valuable for computing infrastructure. GW says the sale will strengthen its finances, with part of the proceeds going toward a new endowment.
The deal has nevertheless raised concerns about what universities could lose when academic property is converted into AI infrastructure. A former GW professor told Fortune that the campus had already experienced declining resources and that faculty and students were not informed beforehand about the potential sale. The broader concern is that universities, under financial pressure from reduced research funding and other challenges, may increasingly view land as a financial asset rather than as space for teaching, research, and innovation.
The University of Michigan represents a different approach. Rather than selling university property to a commercial technology company, it is pursuing a $1.25 billion high-performance computing and research center in Ypsilanti Township with Los Alamos National Laboratory. The university says the facility will support AI and high-performance computing for research in medicine, materials science, clean energy, engineering, and national security. It expects the facility's electricity demand to begin at about 50 megawatts and eventually reach 100 megawatts. Local officials have nevertheless opposed the project.
The larger issue is the growing competition between AI infrastructure and traditional uses of land and resources. Universities need funding and increasingly require powerful computing infrastructure for AI-driven research, while commercial data centers need enormous amounts of land, electricity, water, and connectivity. Fortune's examples show two sides of the same trend: universities can monetize property to strengthen their finances, or they can invest directly in AI infrastructure to expand their research capabilities. Either way, the AI boom is forcing higher education to make difficult decisions about whether its scarce physical resources should serve education, research, financial sustainability, or the rapidly expanding computational economy.
