Data Center Boom: How AI Is Reshaping America’s Construction Landscape
#82: US data center construction soars 190 % since ChatGPT launch – will it surpass office spending?
Are we witnessing a fundamental shift in how America builds its future?
How often do you see a sector go from niche to nearly overtaking a cornerstone of modern infrastructure driving agentic ai within just a few years?
That’s exactly what’s happening in the world of data center construction, as shown by new data released by economist Joseph Politano. In two recent posts, Politano visualises a dramatic transformation underway in the U.S. construction industry—one led by the AI revolution and underpinned by compute expansion + power constraints.
These two critical questions now emerge:
Are we approaching the end of the office‑dominated construction era?
What does it mean when data centers become the centrepiece of physical infrastructure investment in the AI age?
Let’s break down what’s happening, and why it matters for capacity planning.
Last updated: Jan 2026
The U.S. is quietly rewriting its physical economy: office construction is fading while data center construction is exploding. This isn’t just tech growth — it’s an infrastructure shift driven by AI workloads, power constraints, and strategic capacity planning.
What’s happening
+190% data center construction since ChatGPT launch
$40B+ annualized spending (June 2025)
+28% YoY growth
Office construction falling post-COVID
Why it’s happening
In the first chart, Politano highlights how U.S. data center construction spending hit a record high in June 2025, exceeding $40 billion (annualised) for the first time ever. That’s a 28 % increase year‑over‑year, and a stunning 190 % increase since ChatGPT was launched in late 2022. These decisions are not happening in a vacuum; they reflect deliberate infrastructure strategy across the tech sector.
From 2015 to 2020, spending hovered relatively flat, averaging under $15 billion. After the launch of ChatGPT and related ChatGPT-5 setup and tools, however, a sharp and sustained acceleration began—turning a once‑steady line into a near‑vertical growth trajectory. Investors responded to clear usage signals and capacity forecasts, aligning their compute buildout with the explosive demand for AI.
Why the sudden spike? Two key factors:
Explosive demand for AI infrastructure. Large language models and other AI workloads require massive computational resources. This means more GPUs, more servers and more power—all housed in modern, specialised data centers. Supporting these workloads is now a core part of corporate compute strategy.
Cloud computing dominance. As more businesses move operations to the cloud, tech companies are scaling up their physical backbones to support this shift. The decision to double down on cloud capacity reflects a broader shift toward capacity planning in enterprise IT.
Data centers are catching up to office construction
In Politano’s second chart, the broader context comes into view. Just five years ago, office construction spending was seven times larger than data center spending. Office construction peaked around $75 billion, while data centers were below $10 billion.
But look at the crossover trend:
Office construction is in decline, particularly post‑COVID, as remote work has reduced the need for commercial real estate.
Meanwhile, data center construction is in liftoff mode, closing in on parity with office construction as of mid‑2025.
We’re reaching an inflection point where data center construction may soon surpass office construction in total value—something unimaginable just a few years ago. It underscores how much AI‑driven digital transformation has shifted priorities, and it challenges cities and developers to rethink their capacity strategy around physical buildout.
Most AI efforts stall because they never move past the tool layer into a real AI operating model →
The strategist lens: capacity beats ideas
In AI, the constraint isn’t creativity — it’s throughput. The companies that win aren’t the ones with the best slide decks; they’re the ones that secure compute, power, and deployment capacity early.
What it changes (strategic implications)
This isn’t just a story about construction budgets—it’s a broader reflection of how America is redefining the future of work, productivity and infrastructure. Four key implications stand out:
AI infrastructure is becoming a national priority. Compute capacity becomes sovereignty: the winners are the ones who secure AI infrastructure build-out before it becomes politically constrained.
Commercial real estate is being reimagined. Cities will compete on repurposing speed: zoning, permitting, and power access become strategic advantages—not bureaucracy.
Energy demands will skyrocket. Power availability becomes a competitive constraint: winners lock supply early and treat energy like part of their AI stack.
Geopolitical and security considerations. Control of data center footprint becomes leverage: location, ownership, and resilience shift from IT details to board-level strategy.
If AI is “software eating the world”, data centers are the physical layer underneath it. These posts connect infrastructure → strategy → execution.
Build an AI strategy that actually executes
Copy/paste prompt library (strategists)
Agentic systems: what’s coming + what to do
What leaders should do next
As AI continues to embed itself deeper into our lives, data centers will be the backbone of this new era—and the current construction boom is just the beginning. The big question now is:
Will we see a world where data center spending not only eclipses office construction but becomes the dominant category of all non‑residential infrastructure investment?
We’re not far off. As executives craft their next AI infrastructure strategy, they’d do well to consider how the physical footprint of AI will shape markets, cities and energy grids. And as policymakers look ahead, strategic planning (e.g. platform concentration risk) will be essential to balance innovation with sustainability and security.
Hit subscribe to get it in your inbox. And if this spoke to you:
➡️ Forward this to a strategy peer who’s feeling the same shift. We’re building a smarter, tech-equipped strategy community—one layer at a time.
Let’s stack it up.
A. Pawlowski | The Strategy Stack
Very interesting point. I always saw data centers as a crucial part of national sovereignty, but the lens of construction is a nice one. Makes me think about the impacts on architecture and urbanism too.