AI Data Centers Spark a New ‘Man Camp’ Gold Rush

What’s Happening

Here’s a sentence you probably didn’t expect to read this morning: the owner of one of the largest private immigration detention facilities in the United States is pivoting hard into housing for AI data center construction workers. According to a recent report from The New York Times, operators of modular and temporary housing facilities — including those with deep roots in government detention contracts — are eyeing the explosive demand for workforce accommodations near massive rural data center projects. They’re calling them “man camps,” borrowing the term from the oil and gas industry, and the economics are staggering.

The logic isn’t hard to follow. The AI infrastructure buildout currently underway across the United States requires tens of thousands of construction workers deployed to remote and semi-rural locations — places like central Texas, rural Virginia, and the plains of Iowa — where local housing stock simply doesn’t exist. These aren’t weekend projects. A hyperscale data center can take 18 to 36 months to build, and during peak construction, a single campus might need 3,000 to 5,000 workers on-site daily. Those workers need to sleep, eat, and live somewhere. Enter the man camp: temporary, modular housing complexes that can be stood up fast, operated for the duration of a build, and then relocated or repurposed.

What makes this moment especially notable is the crossover from government detention infrastructure. Companies that spent years building and operating facilities designed to house people under government contract — with all the logistical expertise in feeding, sheltering, and managing large populations in remote areas — are now repackaging that exact capability for the private sector. It’s a deeply uncomfortable Venn diagram, and it’s generating both real business momentum and real controversy.

Why Founders Should Care

The numbers here are impossible to ignore. McKinsey estimates that global data center demand will roughly double by 2030, with the U.S. accounting for a disproportionate share of new builds. Reuters reported that Goldman Sachs projects U.S. data center power demand will triple over the same period. Every single megawatt of that capacity requires physical construction — concrete, steel, electrical, plumbing — performed by human beings who need a place to live while they work.

The temporary workforce housing market is already a multi-billion dollar industry, historically driven by oil, gas, and mining. But the AI buildout is creating a parallel demand curve that’s arguably more predictable and longer-lasting than commodity cycles. Unlike an oil boom that can bust overnight when prices crash, the capital commitments from Microsoft, Google, Amazon, and Meta to build AI infrastructure are locked in for years. That translates into a more stable revenue base for housing operators — and a more investable thesis for founders entering the space.

There’s also a timing signal here that matters. We’re still in the early-to-middle innings of the physical AI buildout. Most of the announced data center campuses haven’t broken ground yet. The workforce housing crunch will intensify dramatically over the next 18 to 24 months, and the companies that establish supply now will have significant first-mover advantages in contracts, permitting relationships, and operational know-how.

“The bottleneck for AI isn’t chips anymore — it’s concrete, copper, and places for construction workers to sleep. Whoever solves the workforce logistics problem will capture enormous value.” — Infrastructure investor at a major PE firm, speaking on background

Who’s Already Moving

Management & Training Corporation (MTC) is one of the most prominent names in this story. Long known as an operator of federal prisons and ICE detention facilities, MTC has been publicly exploring workforce housing opportunities tied to large-scale infrastructure projects, including data centers. Their institutional expertise in operating large residential facilities in remote areas gives them a significant operational edge, even as their detention history draws scrutiny.

Target Logistics, a subsidiary of Algeco Scotsman, is arguably the market leader in U.S. workforce housing. They’ve operated man camps for the Permian Basin oil boom, pipeline projects, and disaster recovery efforts. They’ve started aggressively marketing to the data center vertical, and their modular approach — where entire housing villages can be deployed, operated, and relocated — maps well onto the project-based nature of data center construction.

Civeo Corporation, publicly traded on the NYSE, is another established player. Originally focused on mining and oil sands workforce accommodations in Canada and Australia, Civeo has been expanding its U.S. footprint and has signaled interest in the data center construction market as a diversification play. For founders watching public comparables, Civeo’s financials offer a useful template for unit economics in the space.

On the startup side, a handful of smaller companies are emerging. Boxabl, the Las Vegas-based modular housing startup, has attracted attention for its foldable housing units that can be deployed rapidly. While they’ve focused primarily on residential and military housing, their manufacturing approach could scale into temporary workforce applications. Similarly, several venture-backed modular construction startups are beginning to explore data center adjacency as a growth vector.

The Startup Opportunity

For founders, this trend opens several distinct lanes worth exploring:

• Modular and prefab housing for remote deployments. The core physical product — housing units that can be manufactured off-site, shipped to remote locations, and assembled quickly — is ripe for innovation. Current man camp housing is often basic and uninspired. There’s an opportunity to build a better product: more comfortable, more energy-efficient, and designed with worker wellbeing in mind. Happy workers are productive workers, and general contractors know it.
• Workforce logistics platforms. Beyond the physical housing, there’s a massive coordination problem. Who’s tracking which workers need housing at which site, for how long, with what amenities? A SaaS platform that manages workforce accommodation logistics — booking, compliance, billing, occupancy forecasting — could become the “Airbnb for man camps” that general contractors and developers actually need.
• Services and amenities. When you put 3,000 workers in a remote location, you create instant demand for food service, laundry, recreation, healthcare, connectivity, and transportation. Each of these is a wedge for a startup. Think ghost kitchens for construction sites, telemedicine deployments, or rural broadband pop-ups tied to temporary housing installations.
• Sustainability and community relations. One of the biggest friction points in man camp deployments is local opposition. Residents near data center sites don’t love the idea of thousands of transient workers flooding their communities. Startups that help operators manage community relations, minimize environmental impact, and integrate responsibly with local economies will find willing buyers.
• Financing and insurance. Temporary workforce housing projects have unusual financial profiles — high upfront capex, predictable but time-limited revenue, complex insurance needs. There’s room for fintech and insurtech players to build products tailored to this niche.

Watch Out For

Let’s be honest about the landmines here. The optics of this story are brutal. The headline — “detention facility operator pivots to AI worker housing” — practically writes its own critique. Any founder entering this space needs to be aware that the association between detention infrastructure and workforce housing will draw activist attention, media scrutiny, and potential ESG blowback from investors. You don’t have to have anything to do with detention to get caught in the splash zone.

There are also regulatory and permitting risks that shouldn’t be underestimated. Temporary housing facilities are governed by a patchwork of local, state, and federal regulations. Zoning fights can delay projects for months. Environmental reviews, especially near water resources, can be complex. And labor regulations around worker housing — including standards for sanitation, safety, and living conditions — vary significantly by jurisdiction. Bloomberg has reported on growing tensions between rural communities and data center developers over housing impacts, and those tensions will only increase.

Market timing is another concern. While the AI buildout looks durable, construction timelines can shift. If data center developers slow their rollout — whether due to power grid constraints, capital market tightening, or shifts in AI demand — the workforce housing market could contract faster than operators can redeploy their assets. The modularity of man camps helps mitigate this, but it doesn’t eliminate it.

Finally, there’s competition from incumbents. Target Logistics, Civeo, and other established operators have decades of experience, existing client relationships, and operational infrastructure. A startup entering this space needs a genuine differentiation — whether that’s technology, product quality, speed of deployment, or a novel business model — to avoid getting crushed by players who already know how to operate at scale.

The Bottom Line

The AI boom has created a physical infrastructure crisis that most people in tech haven’t fully grasped yet. While the industry obsesses over GPU allocations and model architectures, there’s a massive, unglamorous, highly profitable opportunity in the basic logistics of building AI’s physical backbone — starting with where the workers who pour the concrete and pull the cable are going to sleep at night. It’s not sexy. It’s essential.

If you’re a founder looking at this space, move fast but move thoughtfully. The workforce housing market rewards operators who can deploy quickly, manage complex logistics, and maintain good relationships with local communities and general contractors. The detention-to-data-center pipeline might make for uncomfortable headlines, but the underlying demand signal is real, growing, and largely underserved by technology-forward companies. Build something better than a repurposed detention camp, and you’ll find customers lining up.