Carlyle and KKR win $4 billion U.S. Army contract: Two large data centers to be built at military bases
```
As the Middle East conflict continues, the U.S. Army’s use of artificial intelligence has surged. The U.S. Army has selected private equity giants Carlyle Group and KKR to build data centers at two military bases, with each center estimated to be an investment of about $2 billion. This arrangement marks a new step in the Trump administration’s push for deep integration with the $13 trillion private capital industry.
On March 26, according to the UK’s Financial Times, U.S. Army Secretary Dan Driscoll stated that the Iran conflict has highlighted the urgent need for data centers—the amount of data handled by the Army on AI platforms has surged eightfold in the past few weeks. According to the cooperation arrangement, the private equity institutions will obtain long-term leases at military bases, be responsible for building and operating the relevant facilities, and the Army will not participate directly in the investment, but will receive a certain proportion of dedicated computing power from the data centers.
The report says that both sides are still negotiating and have not signed a final agreement. U.S. Army Deputy Secretary David Fitzgerald revealed that the Army's annual spending on data center use has already reached hundreds of millions of dollars, and future demand is expected to further expand with the spread of artificial intelligence, drones, and other modern war technologies.
Two Data Centers to be Established in Texas and Utah
According to the agreement, Carlyle will build a data center with a capacity of 2.5 to 3 gigawatts on about 1,384 acres at Fort Bliss in El Paso, Texas. The project is expected to start operations at a 200-megawatt scale in 2027 and reach full operational status in 2028.
KKR, leveraging its jointly owned portfolio company CyrusOne with BlackRock, will build a 1-gigawatt data center on about 1,201 acres at Dugway Proving Ground in Utah, projected to begin operations formally in 2029. KKR is one of the world’s largest data center owners.
Under the agreement framework, the Army will receive a certain proportion of dedicated computing power from both data centers, while Carlyle and KKR will sell the remaining computing power to commercial clients, creating a dual military-civilian operating model.
AI Demand Drives Military Infrastructure Transformation; Army Data Center Plans Have Been Brewing for a Long Time
Driscoll stressed in explaining the background of the cooperation that the role of artificial intelligence in modern warfare is rapidly expanding. He noted, “When soldiers face aerial and missile threats from all directions, our demand is huge,” and the computing power needed to deploy drone swarms and launch attacks “requires a different set of infrastructure.”
Although the Iran conflict has further accelerated this demand, Driscoll specifically explained that relevant infrastructure cooperation plans were conceived before the outbreak of hostilities.
Meanwhile, the Pentagon is currently in dispute with AI company Anthropic over terms for using its models. Anthropic’s Claude model is among the earliest known to be used in classified operations, and OpenAI also signed a contract with the Department of Defense earlier this month.
Beyond Data Centers: Army Expands Partnership with Private Equity
Driscoll stated that this cooperation model is expected to be replicated and promoted in other areas of the Army. He suggested that the Army may establish investment vehicles in the future, jointly invest in owning mines with private capital, transport minerals to military bases for processing, and eventually use them to produce weapons components.
This statement indicates that the boundaries of public-private cooperation are expanding far beyond the field of data centers. Fitzgerald and Driscoll both stated that the Army is expected to complete more similar deals.
Risk Warning and DisclaimerThe market involves risks; investments should be made cautiously. This article does not constitute personal investment advice, nor does it take into account the specific investment objectives, financial status, or needs of individual users. Users should consider whether any opinions, views, or conclusions in this article suit their specific circumstances. Investing based on this is at your own risk. ```