特朗普签署行政令限制外国电网设备，AI数据中心工期或承压

Senior News Writer,Data Center Knowledge

August 28, 2026

5 Min Read

President Donald Trump has declared a national emergency over the US bulk power system, directing the Energy Department to restrict certain foreign-produced grid equipment and impose conditions on equipment already in service.

For data center operators racing to secure power for new AI infrastructure, the order could add another constraint to the equipment needed to connect projects to the grid. Transformers, inverters, controls, and other bulk power equipment already face extended lead times and sourcing risks. New federal restrictions could put additional pressure on procurement, project costs and the schedule for bringing new capacity online.

The order comes as utilities and data center developers work to expand generation and transmission infrastructure for large new loads. Its impact will depend largely on how DOE defines covered equipment, suppliers and transactions in rules expected within 120 days.

Trump’s Aug. 26 Executive Order generally prohibits the purchase or installation in the US of certain foreign-produced bulk-power system electrical equipment, including associated critical software and digital capabilities, when it poses specified national security risks. The order focuses on risks including sabotage, unauthorized access and disruption of critical infrastructure. It does not apply to facilities used for local electric distribution.

The White House said the rapid growth of advanced manufacturing, artificial intelligence and defense production has increased US dependence on certain foreign components needed to provide abundant, reliable electricity. Continued reliance on foreign sources of bulk power equipment creates a supply chain vulnerability, the administration said.

The order also applies to equipment already in operation, granting the DOE authority to impose conditions on the continued use and operation of covered equipment. The Energy Secretary must consider reliability, safety, replacement availability, and continuity of service when imposing those conditions.

That could make equipment origin, vendor control, software, remote access, and other lifecycle dependencies part of a reliability determination rather than simply a procurement decision, said Neil Osnato, founder of Persistence Analytics Group.

“The biggest practical change is that equipment origin, vendor control, software, remote access and lifecycle dependencies can now become part of the reliability decision – not just the procurement decision,” Osnato said.

Utilities could have to answer a question beyond whether a piece of equipment works, he said: whether they can continue to rely on it based on who made it, who can access it, what software it depends on and what has changed since it was installed.

Equipment and Software

The order directs the Energy Secretary to identify bulk-power system equipment that may pose the risks outlined in the order and prepare recommendations for addressing those risks. DOE also must publish rules to implement the order.

The scope extends beyond physical grid equipment to associated software, firmware and digital capabilities that could create cybersecurity or operational risks.

“The risk is not limited to the metal sitting in a substation,” Osnato said.

The administration is treating foreign supply chain exposure as a national-security and continuity-of-service issue, rather than simply a trade concern, Osnato said.

The risks include potential cyber or remote-access vulnerabilities inside equipment as well as dependence on foreign suppliers for replacement parts, maintenance, software updates or continued availability, he said.

But determining that risk at the equipment and system level could prove more complicated than identifying where a piece of equipment was manufactured.

“A country-of-origin label alone does not tell an operator whether the full dependency chain is safe,” Osnato said.

NERC said supply chain security is a critical and evolving issue for grid reliability and security.

The North American Electric Reliability Corporation said its Electricity Information Sharing and Analysis Center shares actionable information with industry each day on geopolitical and supply chain security threats targeting the electricity sector.

NERC is also enhancing its supply chain risk management standard and will work with DOE, the Federal Energy Regulatory Commission, the Electricity Subsector Coordinating Council and industry as the administration implements the order, a NERC spokesperson told Data Center Knowledge.

NERC “commends the Administration for the collaborative approach” outlined in the executive order, the spokesperson said.

The order covers a broad range of electric supply chain equipment, including transformers, generators, inverters, batteries, turbines and control systems. Agencies also can consider associated software, firmware, remote-access capabilities, lifecycle maintenance, update mechanisms and other supply chain dependencies.

An August 27, Jefferies equity research note said the order reaches an electricity equipment market already facing elongated lead times and sourcing risks tied to AI and data center development.

The analysts expect the restrictions could contribute to longer lead times and higher domestic pricing, while noting that key implementation details remain unresolved until DOE issues its rules.

The order also creates potential complications for equipment already installed or acquired before it was issued, particularly around servicing, retrofits and replacement. Jefferies said accelerated replacement could further extend lead times.

The order requires DOE to consider reliability, safety, replacement availability and continuity of essential service when determining how compliance is phased in.

Data Center Power

For data center developers and operators, the equipment implications come as power availability and infrastructure timelines have become major constraints on new projects.

Jefferies said the equipment targeted by the order already faces elongated lead times and sourcing risks tied to AI and data center development.

That puts equipment procurement directly into the timeline for bringing new data center capacity online. A transformer or other critical component that cannot be sourced, qualified or replaced on schedule can delay a project even when generation and transmission capacity are available.

Osnato said the challenge will be ensuring supply chain requirements are addressed early enough to avoid problems later in a project.

“The answer should not be to weaken the verification standard,” Osnato said. “It should be to make the evidence standard clearer and earlier in the process so projects do not reach construction or energization before someone discovers that a critical equipment or supply chain assumption cannot be relied on.”

The administration’s implementation rules will determine how the order affects specific equipment, suppliers and projects. DOE has 120 days to publish rules operationalizing the order.

For utilities and data center developers, that leaves an open question over how much additional scrutiny will be required before critical equipment can be ordered, installed and placed into service.

“Announced equipment availability is not the same as executable, reliable equipment availability,” Osnato said.

And after equipment is approved, he said, the next question is: “What changed since it was approved?”

About the Author

Senior News Writer, Data Center Knowledge

Shane Snider is Senior News Writer at Data Center Knowledge, covering AI infrastructure, hyperscale data centers, cloud platforms, and the power and energy systems driving modern compute expansion. His reporting focuses on the operational, economic, and environmental forces reshaping digital infrastructure, including AI factories, utility constraints, liquid cooling, renewable energy procurement, and next-generation data center architectures. He has won recent Azbee awards for news series and government reporting. Based in Raleigh, North Carolina, Snider covers how hyperscalers, utilities, chipmakers, and infrastructure providers are responding to the rapid rise of AI workloads and global compute demand.

You can reach Shane at [email protected] or on LinkedIn.