In the past two years, the AI boom has spilled over into the US power grid. Major tech giants are pouring money into building data centers, but they encountered an unforeseen ceiling—severe shortages of large power transformers.

Demand has long exploded. Order queues can stretch out for years, and production capacity simply can't keep up. But the core issue isn't how slow factory lines are; it's that they can't find enough skilled workers.

You might not know that manufacturing the core components of transformers requires a manual winding process. This isn't a job you can do with the push of a button—training takes 3 to 5 years to become proficient. Factory managers say it's an art that depends entirely on feel and precision; automation can't handle it at all.

Why is there such a labor shortage? Simply put, over the past twenty years, US electricity demand has basically stagnated, and domestic transformer capacity has been shrinking. Now that it's time to use them, they realize their production is far from enough. Statistics show that 80% of large transformers need to be imported, and even half of distribution transformers are reliant on imports.

From another perspective, this bottleneck has a significant impact on global infrastructure investment. Whether it's data centers or other critical industries, rapid expansion depends on sufficient power supply. This current shortfall may take years to fully resolve.