The explosive rise of artificial intelligence is placing unprecedented pressure on power generation, forcing the tech industry to confront hard physical constraints. As advanced AI models expand, the primary bottleneck facing the sector has rapidly shifted from semiconductor supply to raw electrical capacity. On a recent episode of Disruption Interruption, Joe Helfrich, Managing Partner and Co-Founder of NosTerra Venture Capital, explained how this mounting stress on regional power systems could ultimately jumpstart long-overdue innovations in clean energy.

According to projections from the U.S. Department of Energy, national data center energy consumption is expected to double or even triple by 2028, potentially accounting for up to 12 percent of all domestic electricity use. This rapid surge in demand is exposing structural weaknesses in aging transmission networks and generating facilities that were never built for such massive continuous loads. As Helfrich observed, “The demand is still outstripping capacity,” and “the current existing energy grid can’t meet that need.”

While server facilities are frequently criticized as heavy drains on local resources, Helfrich sees a far broader dynamic taking shape. Rather than viewing server facilities solely as grid liabilities, he believes the urgency of the moment presents a rare catalyst for modernization. “Data centers really could be a challenge to energy innovation,” Helfrich notes, while adding, “We take a contrary view and think this is a unique opportunity.”

The challenge will grow even more acute as AI applications transition from massive central training hubs to real-time inference delivered directly to users, devices, and industrial machinery. Performing tasks for autonomous vehicles, medical systems, and complex automation requires immediate response times that centralized facilities cannot deliver. As Helfrich explained, “They’re not going to wait for that inference to go across the country.” Consequently, compute infrastructure must move closer to end users, spreading power demands into new regions and municipal grids.

Solving this dilemma requires looking beyond digital code to the physical foundation of the energy grid. Helfrich emphasizes that real progress depends on an all-of-the-above approach, that spans generation, long-distance transmission, storage, and localized facility management. “It’s going to be an all-of-the-above solution,” he pointed out, emphasizing that “Data centers are really just factories that take energy and convert that into compute.”

Through NosTerra Venture Capital, Helfrich focuses on identifying deep technology solutions capable of making a transformative impact rather than minor operational tweaks. “We look for these mind-blowing, out-of-the-box opportunities,” he said. A compelling example includes an MIT-born startup in the Boston area that embeds carbon-capture systems directly inside modular edge data centers. The technology enhances thermal cooling efficiency while generating beverage-grade CO2 that can be repurposed, and sold to commercial beverage vendors to offset operational costs. Helfrich highlighted this as a prime example of “that crazy out-of-the-box thinking,” capable of “solving two problems at one time.”

Ultimately, the massive capital flowing into AI infrastructure could accelerate energy innovations, that benefit the entire economy. By creating urgent customer demand and reliable revenue streams, data center requirements are unlocking investment for clean energy technologies that previously struggled to scale. As Helfrich emphasized, “This is not about AI,” because “These innovations can make energy more abundant, resilient, and available across the economy.”