Half of US AI Data Centers Delayed as Power Infrastructure Supply Chain Collapses

US AI Build-Out Hits Critical Roadblock

The ambitious push to construct AI data centers across the United States is encountering severe obstacles, with almost half of planned facilities for 2026 expected to be delayed or canceled 1

1

2

2

. The crisis stems from a shortage of electrical components—specifically transformers, switchgear, and batteries—that form the backbone of critical power infrastructure needed to energize these massive facilities 3

3

. Despite tech giants Alphabet, Amazon, Meta, and Microsoft committing to spend more than $650 billion in 2026 to expand AI capacity, neither capital nor ambition can overcome the fundamental supply chain constraints now throttling progress 2

2

5

5

.

Supply Shortage from China Creates Perfect Storm

China has manufactured electrical equipment for US data center construction "for decades," making the supply shortage from China particularly devastating 1

1

. The country accounts for over 40% of US battery imports and maintains nearly 30% market share in certain transformer and switchgear categories 2

2

. Imports of high-power transformers from China surged from fewer than 1,500 units in 2022 to more than 8,000 units through October 2025, according to Wood Mackenzie data cited by Bloomberg 2

2

. During a January visit to a Chinese transformer factory, US utility executives observed that around half the transformers bore US flags, with some specifically destined for data center companies 3

3

.

Tariffs on Chinese Imports Backfire on Trump's AI Ambitions

Donald Trump's aggressive tariffs on Chinese imports are reportedly hindering most data center projects, creating an ironic obstacle to his declared priority of ensuring the US wins the global AI race against China 1

1

. Lead times for high-power transformers have expanded dramatically—what took 24 to 30 months before 2020 now requires wait times up to five years 1

1

2

2

. For AI data centers operating on deployment cycles under 18 months, these delays represent a catastrophe 2

2

. The lag could prove significant, as China is reportedly about five years behind the US in the AI race 1

1

.

Domestic Manufacturing Capacity Falls Short

While Trump would prefer the US manufacture its own equipment, domestic manufacturing capacity for electrical devices "cannot keep up with demand" 1

1

5

5

. Despite a decade of reshoring initiatives, US manufacturing capacity for electrical equipment remains insufficient, forcing AI companies to continue relying on imports even amid tariffs and national security concerns 2

2

. Approximately 12 gigawatts of data center capacity is expected to come online in the US in 2026, according to Sightline Climate, yet only about one-third of that capacity is currently under active construction 2

2

3

3

. "There's not enough domestic capacity to go around, so people are pretty much forced to go to the export market," says Benjamin Boucher, senior analyst with Wood Mackenzie 3

3

.

Community Opposition and Moratoriums Compound Challenges

Beyond supply chain issues, growing community opposition threatens to impose moratoriums on data center construction nationwide 1

1

. Senator Bernie Sanders and Representative Alexandria Ocasio-Cortez announced the Artificial Intelligence Data Center Moratorium Act, which would pause development until safeguards ensure "AI does not increase electricity or utility prices, harm communities, or destroy the environment" 1

1

. Maine is poised to become the first state to pause all data center construction, with legislation likely banning substantial new AI data center construction until 2027 1

1

. At least 10 other states are watching closely, while Denver and Dallas consider similar bans. Smaller cities in Indiana and Michigan have already imposed temporary pauses 1

1

. A Harvard/MIT poll found Americans are more worried about how "giant projects might alter their communities" and quality of life than rising electricity costs 1

1

.