North Carolina’s Data Center Boom 2026
The Race for Power, Investment and the Future of the Digital Economy
North Carolina has spent decades building an economy around banking, manufacturing, technology, research and increasingly sophisticated digital services. Now another industry is arriving at an extraordinary scale.
Data centers.
They are easy to overlook. From the outside, many look like large warehouses surrounded by fencing, substations and acres of land. Inside, however, are thousands of computers working around the clock—the physical machinery behind cloud computing, financial transactions, streaming video, cybersecurity and, increasingly, artificial intelligence.
And North Carolina is becoming part of a much larger national race. The state currently has 13 operating data centers representing approximately 506 megawatts of capacity, according to Cleanview. Another 52 projects representing approximately 3,975 megawatts are currently planned. If every planned project were ultimately built at its listed capacity, North Carolina would move from roughly 506 MW of operating capacity to nearly 4,481 MW of operating and planned capacity combined.
That is a remarkable expansion.
But there is an important distinction between planned and operating.
A data center can be announced long before its first server is switched on. Between those two points are land acquisition, financing, zoning, permits, transmission, substations, generation capacity, water infrastructure, equipment and an agreement with the utility to actually deliver the electricity.
That makes North Carolina's data-center story bigger than a story about buildings.
It is a story about electricity, water, taxes, infrastructure, economic development and who ultimately pays for growth.
North Carolina Is Entering the Race From a Different Starting Point
North Carolina is not Virginia. It is not Texas. Those states already have substantially larger operating data-center markets. Cleanview's September 2026 data illustrates the difference:
State | Operating Capacity | Planned Capacity |
Virginia | 17,325 MW | 34,578 MW |
Texas | 7,169 MW | 96,993 MW |
Ohio | 4,121 MW | 23,504 MW |
Georgia | 2,968 MW | 16,466 MW |
Arizona | 3,665 MW | 12,831 MW |
Utah | 1,184 MW | 31,603 MW |
North Carolina | 506 MW | 3,975 MW |
Cleanview, September 2026.
The numbers tell two stories simultaneously.
Virginia has already built an enormous data-center ecosystem. Texas has an extraordinary development pipeline. Georgia and Ohio have established operating markets with substantial additional construction planned.
North Carolina, by comparison, has a relatively small operating base—but its planned capacity is nearly eight times its existing operating capacity. That is the number worth watching. North Carolina is not simply adding another data center here and another there. It is potentially entering the next stage of its digital infrastructure economy.
The Real Race Is Not to Announce a Data Center
There is a temptation to count announced projects. That can be misleading. A better way to understand the industry is to think of development as a pipeline:
Announced → Permitted → Power Secured → Under Construction → Operational
Every step removes another uncertainty. This matters because the national data-center pipeline has become enormous. Cleanview currently tracks approximately 61.5 gigawatts of operating U.S. data-center capacity and another 379.8 gigawatts of planned capacity.
Obviously, not every planned megawatt will become an operating megawatt.
Some projects will be delayed.
Some will be redesigned.
Some will lose financing.
Some will encounter transmission constraints.
Others may never be built.
The states that succeed in this next phase will therefore not necessarily be the states with the largest announcement pipelines. They will be the states capable of turning proposed projects into physical infrastructure without creating unacceptable costs or reliability problems elsewhere in the system.
Virginia
The Established Giant
Virginia provides North Carolina with perhaps the clearest example of what happens when data-center development reaches enormous scale. Cleanview identifies 352 operating facilities in Virginia with 17,325 MW of operating capacity, plus another 391 planned facilities representing 34,578 MW. Northern Virginia has become one of the world's most concentrated data-center markets.
But success creates its own problem.
Once enough facilities arrive, electricity demand becomes an infrastructure issue rather than merely an economic-development opportunity. Virginia has responded by creating new rules for large electricity customers. Beginning in 2027, new large-load customers contracting for service will face minimum obligations associated with transmission and distribution costs. Virginia's State Corporation Commission says large-load customers will be required to pay at least 85% of the transmission and distribution costs incurred to serve them each month, subject to the commission's rules and exemptions. Virginia is therefore attempting to answer a difficult question:
How do you continue attracting enormous private investment without allowing the cost of serving those customers to become someone else's problem?
That question is coming to North Carolina.
Texas
Build at Scale, But Verify the Load
Texas represents another model. The state has 139 operating data centers representing 7,169 MW, according to Cleanview. But the planned pipeline is extraordinary: 363 planned projects representing nearly 97,000 MW. That figure should not be interpreted as 97 GW that Texas will necessarily build. It is a development pipeline. And Texas regulators are now taking a closer look at what is actually behind those requests.
In August 2026, Governor Greg Abbott directed the Public Utility Commission of Texas and ERCOT to conduct a comprehensive audit of data centers moving through the state's interconnection process.
The review includes projected electricity demand, on-site generation, water requirements, cooling systems, incentives and other factors. Projects that fail to meet the requirements are subject to denial of connection to the Texas grid.
The scale of the problem is striking. ERCOT was considering more than 474 GW of requests to connect to the Texas grid, according to the governor's office—more than five times Texas' record peak electricity demand. Texas is essentially asking developers to answer a very practical question:
You want the electricity. Show us how the project will actually be served.
Georgia
Make Large Loads Carry More of the Cost
Georgia offers another model. In January 2025, the Georgia Public Service Commission approved rules allowing Georgia Power to impose additional terms on new customers using more than 100 MW. Those customers can be required to pay costs associated with upstream generation, transmission and distribution as construction progresses. The rules also allow longer contracts and minimum billing requirements. The reasoning is straightforward.
A utility may build infrastructure specifically because a massive new customer says it needs it. If that customer subsequently reduces consumption or leaves, somebody still owns the infrastructure. Georgia's approach attempts to make sure that the financial obligation does not simply migrate to ordinary residential and commercial customers. Georgia has simultaneously been planning substantial additional generation to meet expected demand. Again, the lesson for North Carolina is less about data centers themselves than about cost allocation.
Arizona
Protect the Existing Ratepayer
Arizona is confronting a similar challenge. The Arizona Corporation Commission has been examining the rapidly growing demand from data centers and other large-load customers. In April 2026, the commission noted that utilities were receiving interest from large customers beyond their current ability to serve them. The commission's discussion centered heavily on ensuring that new growth pays for the demands it places on the electric system.
That is an important principle. Economic development can produce enormous private investment. But electricity infrastructure is often a public utility system.
The question becomes:
How much of the infrastructure required for a private project should be paid by the project, and how much should be recovered from the broader customer base?
There is no universal answer.
But states are increasingly recognizing that the question needs to be answered before the infrastructure is built.
Ohio
Put the Grid Connection Under the Microscope
Ohio is considering an even more direct approach. Senate Bill 381 would require Public Utilities Commission of Ohio approval of agreements to connect data centers to the electrical grid. The legislation was introduced in March 2026 and referred to the Senate Public Utilities Committee. The proposal illustrates the growing importance of the interconnection process.
A developer can own land and have financing.
It can have customers.
It can have a building permit.
None of that guarantees that the electric grid can safely provide hundreds of megawatts of continuous power.
The interconnection agreement is where the proposed facility meets the physical reality of the electric system.
Utah
Is Watching Something Else - Water
Electricity is only half of the infrastructure equation. The other increasingly important resource is water. Utah enacted a new law in 2026 requiring large new data centers—generally facilities larger than 10,000 square feet intending to withdraw at least 75 acre-feet of water annually—to notify the state's Division of Water Rights and report projected water use before construction.
Once operational, facilities must report actual water withdrawals and efforts to reduce consumption annually. Utah's approach is notable because it turns water consumption into something that can actually be measured publicly. That may sound mundane.
It is not.
If a community is deciding whether to approve a major industrial facility, knowing how much electricity and water it will consume is basic information. North Carolina's own Energy Policy Task Force has recognized this gap.
North Carolina
Is Still Building the Framework
North Carolina's government is not ignoring the issue. Governor Josh Stein established the North Carolina Energy Policy Task Force in August 2025 specifically to examine the state's growing electricity demand, including demand from data centers and other large industrial users.
The task force's February 2026 interim report identified a series of recommendations:
Develop large-load electricity tariffs.
Explore "bring your own capacity" arrangements.
Encourage load flexibility.
Reform large-load interconnection.
Evaluate the value of existing data-center tax exemptions.
Examine advanced transmission and grid-enhancing technologies.
Develop third-party load forecasting.
Explore energy and water-use reporting for data centers.
That last point is particularly revealing. The task force found that North Carolina does not currently have reliable public data on the total number and size of existing and proposed data centers in the state.
Think about that for a moment.
North Carolina is preparing for substantial electricity-load growth from data centers while simultaneously acknowledging that the public data describing the state's data-center footprint is incomplete.
It is a sign that the industry has grown faster than the state's information infrastructure. And it presents an opportunity.
The North Carolina Question: Who Pays?
There is a natural attraction to a billion-dollar data-center announcement. Construction workers get jobs, Developers spend money, Land values can rise, Counties can receive additional property-tax revenue, Utilities sell more electricity, Technology companies gain infrastructure.
The state becomes more attractive to companies building the next generation of cloud and AI systems. Those benefits are real. But so are the costs.
A large data center may require:
Transmission lines.
Substations.
Generation capacity.
Road improvements.
Water infrastructure.
Fiber infrastructure.
Emergency services.
Land.
And sometimes substantial public incentives.
That creates the question at the heart of the North Carolina story:
When a community welcomes a large data center, how much of the economic benefit stays with the community, and how much of the infrastructure cost follows the project?
There is no reason that question needs to be hostile to development. Quite the opposite. A predictable, transparent cost structure can make development easier because developers know the rules before they commit capital.
Charlotte Has a Particularly Interesting Position
The Charlotte region deserves special attention.
Charlotte is one of the country's major financial centers, with a large concentration of banks, investment firms, insurance companies, fintech companies and professional-services organizations.
Those businesses increasingly depend upon cloud infrastructure.
The physical data center may be dozens of miles away from the corporate headquarters, but the economic activity it supports can be deeply connected to Charlotte. Cleanview currently lists a planned 400-MW Digital Realty Charlotte Campus in Mecklenburg County. It also lists multiple PowerHouse projects in the Charlotte market, each at approximately 75 MW.
That creates an interesting economic relationship:
Charlotte's financial economy creates demand for digital infrastructure, while digital infrastructure increasingly becomes part of the economic foundation supporting Charlotte's financial economy.
For a state already known for banking and financial services, that connection matters.
The Human Cost—and the Human Opportunity
It is easy to discuss data centers entirely in megawatts. But megawatts do not live in communities—people do. A family sees the electric bill. A farmer sees the transmission line. A homeowner sees the new industrial campus down the road. A county commissioner sees the property-tax revenue. A utility customer sees a rate case. A construction worker sees a paycheck. A small business sees a new customer base. A water manager sees another demand on the watershed. An economic-development official sees billions of dollars of potential investment.
All of those perspectives can be true at the same time. That is why the debate over data centers should not be reduced to whether they are "good" or "bad." The more useful question is whether the economics of each project have been properly understood.
Planned Is Not the Same as Built
This may be the most important number for North Carolina to watch over the next several years.
Today, the state has approximately: 506 MW operating versus 3,975 MW planned. That creates a planned-to-operating ratio of almost 8 to 1. But the same phenomenon appears elsewhere.
Texas has roughly 7.2 GW operating and nearly 97 GW planned.
Utah has only 1.2 GW operating but more than 31.6 GW planned.
Arizona has 3.7 GW operating and 12.8 GW planned.
Ohio has 4.1 GW operating and 23.5 GW planned.
Georgia has 3.0 GW operating and 16.5 GW planned.
Virginia, meanwhile, has already built 17.3 GW and has another 34.6 GW planned.
The lesson is simple:
The size of a state's pipeline tells us what developers want to build. It does not tell us what will actually get built.
The conversion from planned to operational capacity may ultimately become the more meaningful measure of the data-center race.
A New Way to Measure the Data Center Boom
For North Carolina residents, investors and policymakers, a useful data-center dashboard should eventually track five stages:
Announced - A developer has publicly identified a proposed project.
Permitted - The project has received the necessary governmental approvals.
Power Secured - The developer has established how the facility will obtain reliable electricity.
Under Construction - Physical construction has begun.
Operational - The facility is actually consuming electricity and serving customers.
That framework would give North Carolinians something more valuable than a list of announcements.
It would show progress.
And it would allow communities to compare promises with reality.
North Carolina's Opportunity
North Carolina has a choice to make—not about whether the digital economy is coming, but about how the state accommodates it. The data-center industry is not disappearing: artificial intelligence, financial services, cloud software, healthcare, and increasingly government all require computing infrastructure. The question is how North Carolina builds that infrastructure while maintaining reliable electricity, protecting water resources, preserving reasonable utility costs, and ensuring that communities receive a fair economic return.
Other states are already experimenting with different answers. Virginia is tightening the economics of large-load customers. Texas is auditing projects before they receive grid connections. Georgia is requiring large customers to assume more of the infrastructure costs created by their demand. Arizona is examining ratepayer protections. Ohio is considering greater regulatory oversight of grid connections. Utah is requiring greater transparency around water consumption. North Carolina is now developing its own framework, which makes this moment particularly important—the state still has time to learn from the successes, and the problems, of the markets that moved first.
The Bottom Line
North Carolina's data-center boom is not really about computers. It is about power, water, land, taxes, and infrastructure—and ultimately, it is about the economic choices a community makes when billions of dollars of private investment arrive at its doorstep. The state has approximately 506 MW of operating data-center capacity today and another 3,975 MW in the current planned pipeline. Some of that pipeline will become reality, and some will not; the important thing is that North Carolina knows the difference.
Because the next chapter of the state's data-center story should not be measured by how many projects are announced. It should be measured by how much value those projects create, what infrastructure they require, and whether the people who live around them share fairly in the benefits of the digital economy they are helping to build.
That is the real story behind North Carolina's data-center boom.
Comments