Why AI’s Power Problem Is Higher Ed’s Too
by Claire Brady, EdD
“One data center can use as much power as 1,000 Walmart stores, and AI training consumes ten times more energy than a standard internet search.”
Last week’s Wall Street Journal story could have been pulled straight from a science-fiction plot: major tech companies are now building their own power plants to keep up with the energy demands of artificial intelligence.
In West Texas, OpenAI and Oracle are constructing natural-gas facilities to power their $500-billion Stargate project. Elon Musk’s xAI is doing the same in Memphis. Equinix, which operates data centers across the U.S., is relying on fuel cells. And Meta is planning to bypass the electrical grid altogether for one of its new campuses in Ohio.
The reason? The U.S. power grid simply can’t keep up. The surge in AI computing requires electricity on a scale the system wasn’t designed for. One data center can use as much power as 1,000 Walmart stores, and AI training consumes ten times more energy than a standard internet search.
For those of us in higher education, this story is about more than turbines and transmission lines—it’s about foresight, stewardship, and readiness.
Why It Matters for Colleges and Universities
1. Energy and infrastructure are becoming academic issues.
AI adoption isn’t just a technology conversation—it’s an infrastructure one. As campuses scale up their use of AI in research labs, admissions modeling, student advising, and administrative automation, the computing load will grow exponentially. Each new use case compounds the demand for electricity, cooling, and network stability. Institutions that fail to integrate energy forecasting into their digital strategy could soon discover that the limits of innovation aren’t financial or regulatory—but physical. Facilities and IT leaders will need to collaborate more closely than ever, treating data and energy as interconnected assets.
2. Sustainability commitments will be tested.
Many colleges proudly display countdown clocks to 2030 carbon neutrality goals. Yet the rise of energy-intensive technologies like AI threatens to widen the gap between aspiration and action. True sustainability in the AI era will require new frameworks that balance innovation and responsibility: green computing standards, energy-efficient data management, and partnerships with renewable providers. The institutions that succeed will position AI not as a sustainability setback but as an accelerant for smarter, cleaner systems—from optimizing HVAC and transportation to managing microgrids and campus energy dashboards.
3. Location and partnerships will matter more than ever.
Just as access to railroads and ports once determined economic opportunity, proximity to reliable and affordable energy will soon shape institutional competitiveness. States like Texas, Oklahoma, and Ohio are already courting tech companies with streamlined permitting and access to renewables or natural gas. For higher ed, that same geography could define research competitiveness and industry collaboration. Universities in these emerging “AI energy corridors” might find themselves better positioned for grants, data-sharing partnerships, or access to next-generation cloud and compute resources.
4. Equity and access remain essential lenses.
As data centers compete for limited power and utilities prioritize industrial clients, colleges and communities could face tough choices. How do we ensure that the power driving AI innovation also sustains educational equity and local well-being? This is not only a budget question—it’s an ethical one. Higher education has a responsibility to advocate for fair access to energy and to model what shared benefit looks like in a resource-constrained future. Our decisions about what gets powered—and why—should always circle back to mission, students, and social responsibility.
Leading Through the Transition
Campus leaders don’t need to start building power plants—but they do need to think like systems designers. This moment calls for collaboration among facilities, IT, sustainability, and academic affairs. It may also require updating procurement language, exploring renewable power agreements, and building awareness among faculty and students about the real costs of AI adoption.
AI is reshaping the educational landscape, but also the physical one. Power, once taken for granted, is now a strategic resource. The most forward-thinking institutions will connect technological ambition with environmental responsibility and operational realism.
The question for higher ed isn’t whether we can keep up with AI’s energy appetite—but how we can lead with intention while the world races to keep the lights on.
Five moves to make now
Run a dual audit: electrons and ethics. Pair a facilities-level load and cooling assessment with an AI roadmap that prioritizes student impact, research value, and equity. If power is scarce, what deserves it most?
Revise your RFPs and MOUs. When negotiating with cloud providers, colocation partners, or AI vendors, add clauses on power provenance (renewables content), demand flexibility, and emissions reporting.
Stand up a Resilient Compute Task Force. Bring together IT, facilities, sustainability, finance, and academic leadership to scenario-plan: on-site generation, heat reuse, battery storage, and demand response.
Pursue portfolio power. Blend long-term clean PPAs with campus solar + storage, evaluate waste-heat capture from data rooms, and explore community microgrids that support critical campus functions during outages.
Teach the moment. Build energy and data-center literacy into engineering, policy, and business curricula. Students should understand why “AI equals energy” and how design choices—from model size to scheduling—influence emissions and cost.
Read the full article here: https://www.wsj.com/business/energy-oil/ai-data-centers-desperate-for-electricity-are-building-their-own-power-plants-291f5c81?st=STmBzj&reflink=desktopwebshare_permalink
