Commentators debating whether artificial intelligence is a bubble often look to the dotcom boom as the obvious comparison. That analogy may help explain valuations, exuberance, and the likelihood that some companies will fail. But it misses the scale of what is happening beyond the technology sector.
A more useful comparison is the Industrial Revolution, which I see as the last technological shift that reorganized the physical economy. Steam power did not simply create new machines or businesses. It changed where factories were built, how goods moved, and which cities became centers of economic activity.
AI is beginning to create a similar dynamic. Its most visible products may be digital, but its impact will be physical. Its growth will depend on an enormous buildout and on which places are poised to deliver it.
Land, power, and financing are necessary, but not sufficient. The regions that capture the next wave of AI-driven investment will also need grid capacity, water access, connectivity, permitting, labor, and community support. Increasingly, the competitive advantage lies in the ability to align those pieces and secure permission to build.
THE WRONG ANALOGY
The dotcom framing encourages investors to ask which AI companies will win. The Industrial Revolution framing asks a more consequential question: What physical systems must be built and where for AI to become embedded across the economy?
AI requires data centers, power generation, grid capacity, cooling systems, fiber networks, and developable land. The IEA projects that global data center electricity consumption will roughly double to around 950 terawatt hours by 2030. McKinsey estimates the sector may require $6.7 trillion in global capital expenditure by the end of the decade. Those numbers are enormous, but the more interesting story is what happens around that investment.
As demand for computing power grows, it will influence where businesses expand, where utilities invest, and which regions can accommodate economic activity. It may change the value of land, the importance of energy infrastructure, and how investors assess industrial, logistics, and other real assets.
Companies have traditionally made location decisions based on access to customers, talent, transportation, and cost. Those factors still matter, but so do power, water, and the ability to secure approvals on a commercially viable timeline.
PERMISSION TO BUILD IS BECOMING A COMPETITIVE ADVANTAGE
The Industrial Revolution created enormous prosperity, but it also showed the costs of building without sufficient regard for communities and resources.
Today, communities are already asking harder questions about water use, grid reliability, jobs, and local economic benefits. As AI infrastructure expands, those questions will become economic as much as political. A project can have land, financing, and even access to power and still fail if it does not have the support required to move forward. In that sense, public trust is becoming another form of infrastructure.
Projects that solve a technical constraint while creating new strain for the surrounding community will face resistance, delay, and reputational risk. Community engagement, resource planning, and shared economic benefits must be built in from the start, not added later.
This is not an argument for building less. If AI is going to deliver the productivity gains being promised, economies may need to build an extraordinary amount of infrastructure. The challenge is building it at the speed and scale required while creating tangible benefits for communities absorbing that growth.
AI COULD CHANGE THE MAP OF ECONOMIC VALUE
In the AI economy, the definition of a strong location is being rewritten. Industrialization in the 19th century pulled populations toward energy, logistics, and production hubs. AI will not replicate that transformation, but it is introducing a powerful new variable into economic geography: access to the physical resources required for computing at scale.
This changes the role of real assets. They are no longer simply the container for economic activity. Increasingly, they are part of the operating system that makes that activity possible. Land, energy, infrastructure, and the built environment must be planned together.
That requires closer coordination among developers, utilities, technology companies, public authorities, and capital partners. The challenge is no longer simply finding the right site. It is assembling a system capable of delivering site, power, infrastructure, and approvals at the right time.
The opportunity has expanded beyond owning a data center, power asset, or tract of land. Value may increasingly accrue to places and platforms that can solve multiple constraints at once.
THE BUILDOUT REQUIRES A DIFFERENT KIND OF CAPITAL
The Industrial Revolution depended on enormous private investment alongside public institutions. Today’s capital markets are vastly different, but technological transformation still requires investment in physical systems before its full economic benefits can be realized.
Pension funds, sovereign wealth funds, and individuals whose savings are invested through public and private markets can help finance the physical foundations of AI. For long-term investors, that may create an opportunity extending beyond identifying the next technology winner, touching power, industrial property, logistics, housing, and communities.
Private capital will not deliver it alone. Utilities, local authorities, and public institutions will shape energy, land use, and development policy, making coordination between private capital and public stakeholders critical to where and how quickly this buildout happens.
Capital may be abundant. The ability to deploy it into projects that can be delivered is much scarcer.
THE AI ECONOMY WILL BE BUILT ON THE GROUND
The most important AI investment may not be an AI company. It may be the power system, industrial site, logistics network, workforce ecosystem, or community partnership that moves the technology from experimentation into everyday economic life.
We spend significant time asking which companies and models will win the AI race. That matters. But technological revolutions are not ultimately defined by the technology alone. They are defined by what economies build around it.
AI may live in the cloud. But the leaders of the AI economy will be determined on the ground.
David Steinbach is global chief investment officer at Hines.