Matt Champlin / Getty Images
In 1877, mining mogul George Hearst bought a gold claim in the Black Hills of Dakota Territory for $70,000. Within two years, he had consolidated neighboring claims, monopolized local water rights, and built a railroad to connect his holdings to eastern markets. The land on which he built his empire had been guaranteed to the Sioux under the Fort Laramie Treaty of 1868, only to be seized after gold was discovered and the U.S. government reneged on the agreement.
Today, a different kind of extraction is reshaping the American map. In an AI-driven data center land rush, developers are acquiring rural parcels not based on what the land can grow or who will live there, but on how many megawatts the grid can deliver. The mechanics differ. The underlying pattern does not.
The first American land boom built on this template was the transcontinental railroad. Congress authorized four transcontinental railroads and granted 174 million acres of public land for rights-of-way, beginning with the Pacific Railway Act of 1862, which provided federal land grants and loans for construction.
The scale was staggering. From 1850 to 1871, the railroads received more than 175 million acres of public land, an area larger than Texas. Construction preceded settlement: railroads were built across the Plains before communities existed, and the rail lines themselves shaped the placement and layout of towns.
The structure of these grants embedded a logic that would repeat in later booms. Capital moved first, backed by government subsidy, and governance trailed behind. Anti-monopolists argued that railroad monopolies originated not from efficient investment but from special privileges afforded by the government: land grants, loans, bonds, and political contributions that secured supporters in legislatures, Congress, and the courts. As stronger railroads bought up weaker companies and divided up markets, the dangers of monopoly became more apparent.
The consolidation was swift. The Northern Securities Company, formed in 1901, combined the interests of J.P. Morgan, James J. Hill's Great Northern Railway, the Northern Pacific Railway, and the Chicago, Burlington and Quincy Railroad, establishing effective monopoly control over rail transportation in the northwestern U.S. It was not until the Sherman Anti-Trust Act of 1890 that federal law began to reverse the trend toward unchecked consolidation that had accelerated after the Panic of 1873.
The western mining booms operated on similar principles, with even less pretense of governance. According to the Congressional Research Service, the primary purposes of the General Mining Law of 1872 were to "promote mineral exploration and development on federal lands" and "help settle the West". The law set the price of a mining claim at $2.50 to $5.00 per acre, a price that has remained unchanged since 1872. The law contained no royalty provisions, and critics noted that billions of dollars in federal resources could pass into private hands for a pittance.
In the Black Hills, the capital moved with open disregard for existing legal agreements. In 1874, General George Custer led an expedition to search for gold in the Black Hills; by 1875, some 800 miners had flooded into land reserved by treaty exclusively for the Sioux, and a U.S. decree confining Lakota, Cheyenne, and Arapaho peoples to the reservation violated the Fort Laramie Treaty of 1868.
The consolidation that followed tracked the railroad pattern. George Hearst spent two years in Lead securing control of the Homestake and neighboring mines. By 1900, Homestake owned 300 claims on 2,000 acres. His company grew to be the largest private mining firm in the United States.
"An often-overlooked principle of American capitalism is its tendency toward monopolistic forms of production," historian John Reeves recently told Quartz. That was "certainly the case" in the 19th century. "And we see unregulated (more or less) monopolies like Amazon $AMZN, Meta $META, and Google $GOOGL dominating everywhere today."
The Federal-Aid Highway Act of 1956 offered a mid-century iteration. With an original authorization of $25 billion for the construction of 41,000 miles of interstate highway, it was the largest public works project in American history at the time.
The language of progress and national interest framed the project. President Dwight Eisenhower called the highway system "essential to the national interest." But the costs fell unevenly. The interstate highway system displaced minority neighborhoods in urban centers: between 1957 and 1977, the system displaced more than 475,000 households and 1 million people across the country, while creating physical barriers between neighborhoods and reducing the availability of housing.
The affected communities were rarely consulted before a development project that would impact their fate broke ground. Research has found that redlined neighborhoods were three times more likely to have an interstate highway routed through them compared to other neighborhoods, according to the U.S. Environmental Protection Agency.
The victims of highway building were overwhelmingly poor and Black, and a general pattern emerged of using highway construction to eliminate "blighted" neighborhoods and redevelop valuable inner-city land.
The AI-driven land rush shares structural features with each of these episodes. Capital moves ahead of governance: local governments use nondisclosure agreements to keep data center development secret, with officials in Minnesota, Virginia, and other states aware of proposals for months or years before disclosing them, according to Wisconsin Watch. In Beaver Dam, Wisconsin, Meta used two shell companies to develop a $1 billion project in secret.
The language of inevitability is familiar. Data center proponents tout community benefits and job creation, even as the primary requirements are electricity, water, and land. Reeves pointed to the same rhetoric accompanying westward expansion: Settlers understood that land seizures violated treaties, but many believed economic development could not and should not be stopped.
And consolidation follows the scramble. According to estimates from Synergy Research Group, Amazon's share of the worldwide cloud infrastructure market was 28% in the first quarter of 2026, with Microsoft $MSFT Azure at 21% and Google Cloud at 14%; together, the three account for more than 60% of the market. Hyperscale operators controlled 44% of global data center capacity in 2025, up from 20% in 2017, with the share projected to reach 61% by 2030, according to Synergy Research Group data compiled by Cargoson.
Data center mergers and acquisitions reached $57 billion in 2024, and 2025 saw the sector's largest deal in history when a consortium including BlackRock $BLK, Global Infrastructure Partners, and Abu Dhabi's MGX acquired Aligned for $40 billion, according to Data Center Dynamics.
The parallels are not perfect. Data centers do not require the seizure of treaty-protected land, and the interstate highways were a direct government project, not a private scramble facilitated by public subsidy. But the structural similarities hold. In each case, a new technology or strategic priority made certain land valuable in ways existing residents and institutions did not anticipate. In each case, decisions about who benefits and who bears the costs were made before the public had time to organize. In each case, the initial rush gave way to consolidation by a small number of dominant firms.
The question raised by this history is not whether the data center boom will reshape American geography. It already is. The question is whether this time, governance catches up to capital before the map is redrawn.
