I came to energy from telecom, and the parallels between the two industries are hard to miss. Both started as regulated monopolies within a few years of each other in the late 1800s. Both promised universal service at low residential rates. Both built business models that rewarded heavy physical infrastructure above all else.
Telecom broke out of that model. The power industry is only beginning to.
In the 1990s, mobile phones ran into what looked like a dead end. Emergency 911 service depended on landlines at fixed, known addresses. To regulators, a network with dropped calls and callers they could not locate looked unsafe for emergencies, which seemed to cap how far wireless could go.
Carriers answered with enhanced 911 location technology, and regulators signed off. Engineering alone didn't get them there. Companies changed how they earned revenue, customers changed what they paid for, and the rules governing the industry changed with them. Wireless went from novelty to normal in about a generation.
Power has its own fixed assumption: generation must follow load. For a century, utilities have kept enough dispatchable plants on hand to meet any peak whenever it arrives, meaning gas, coal, nuclear and hydro. Regulation and business models reward exactly that kind of investment.
Pressure on that assumption is building. Home electricity rates climbed almost 5% last year, and data centers are adding load faster than utilities can plan for it. Solar and wind now provide the cheapest new power and nearly all new capacity, yet neither can ramp up on command. They work best alongside batteries or demand that moves to meet them.
Utilities have leaned on flexible demand for a long time. American factories have cut back at peak hours for decades. Australia has timed household water heaters since long before smart homes. Virtual power plant firms like Leap and Virtual Peaker now pay homes and businesses to shift use through utility programs.
Data centers could do the same at far larger scale. Researchers at Duke University's Nicholas Institute estimated in February 2025 that if data centers trimmed use for a small share of their hours each year, the current grid could take on about 100 gigawatts of new load without major new plants. A later study found that a 1% to 2% cut in data center peak demand could lower retail rates by 0.5% to 2.8%.
Telecom moved on three fronts at once: technology, business models and regulation. Power needs the same. One difference makes it harder. People adopted wireless by buying phones. The grid's shift depends on large customers and ratepayers pressing utilities for affordable, flexible power.
Telecom's turn began when its hardware went digital. Merge4 brings that shift to power distribution, with silicon carbide transistors in place of mechanical equipment, protection in microseconds and intelligence at every node. We think power is ready for a turn of its own.
Anna is a serial entrepreneur, CEO, and strategist working at the intersection of AI, energy, and critical infrastructure. She holds a Ph.D in smart grid technology and writes for Forbes on AI, electrification, and the modernization of energy systems.
