The global artificial intelligence boom could trigger one of the largest technology infrastructure investment cycles in history, with $31.6 trillion in capital spending potentially flowing into data centers and related computing infrastructure through 2050, according to a new report from PwC.

The estimate highlights how quickly AI is changing the technology industry. As companies deploy increasingly powerful AI models, demand is rising for data centers, processors, networking equipment, storage and electricity needed to operate them.

PwC's forecast puts annual data center capital spending at approximately $800 billion in 2026, rising to about $1.1 trillion by 2030 and reaching $1.8 trillion annually by 2050. The long-term figure could approach $50 trillion if AI adoption grows faster than expected.

AI Is Creating a Different Infrastructure Cycle

The scale of the projected investment is significant, but the way the spending will occur may be even more important.

Traditional infrastructure projects generally require a large initial investment followed by years of maintenance. AI infrastructure is different because much of the technology inside data centers becomes outdated quickly.

Servers, GPUs, networking equipment and other computing hardware may require major upgrades every four to six years. That means investment will continue even after the physical buildings have been completed. PwC estimates that technology equipment will account for an increasing share of total investment, reaching about 93% by 2050, compared with roughly 70% today.

This creates a long-term spending cycle rather than a one-time construction boom.

Power Is Becoming a Critical Issue

The biggest limitation may not be the availability of land or computer chips. It may be electricity.

AI data centers require enormous amounts of power, particularly as companies operate increasingly large models and expand AI services to millions of users.

PwC identifies access to affordable, reliable and low-carbon electricity as the most important factor determining where future AI infrastructure will be built. Connectivity, regulation, security and access to GPUs will also influence investment decisions.

That is already changing where data center developers look for new sites.

In Europe, for example, developers are increasingly considering locations farther from major cities because those areas can offer more available land and electricity. The shift demonstrates how access to power is becoming a central factor in the geography of the technology industry.

The Investment Will Extend Beyond Technology

The AI infrastructure boom is also creating demand in industries outside traditional technology.

New data centers require land, buildings, electrical connections, cooling systems, fiber networks and specialized construction. They also require power generation and transmission infrastructure capable of supporting large facilities.

That makes AI increasingly connected to the construction, energy, real estate and financial sectors.

The U.S. is expected to capture nearly half of the projected global investment, with PwC estimating approximately $15.1 trillion in cumulative spending through 2050. Asia-Pacific is projected to receive about $8.2 trillion, led by major technology markets including China and India.

The competition for this investment could encourage governments and local communities to accelerate permitting, expand power capacity and improve digital infrastructure.

Data Centers Are Already Affecting the Economy

The long-term forecast comes as the current data center construction boom is already having a noticeable effect on the U.S. economy.

Construction spending on data centers reached an annual rate of approximately $75 billion in July, up 57% from a year earlier, according to recent economic data. The expansion is increasing demand for construction workers, materials, electricity, computer equipment and related services.

At the same time, the rapid growth is creating challenges for electricity providers.

Some U.S. utilities have received requests for enormous amounts of new power capacity from proposed data centers. Regulators in states including Texas have begun questioning whether all of the proposed projects are realistic, raising concerns about speculative demand and the cost of expanding electricity infrastructure.

That tension could become more important as AI companies continue building larger computing facilities.

A Massive Opportunity With Significant Risks

The projected investment represents a major opportunity for technology companies, utilities, construction firms, semiconductor manufacturers and infrastructure investors.

Companies producing AI processors and networking equipment are already benefiting from the spending wave. Broadcom, for example, recently raised its forecast for AI chip revenue to approximately $115 billion in fiscal 2027, with the company expecting that figure to reach about $230 billion in 2028.

But the size of the investment also creates risks.

If AI adoption grows more slowly than expected, some data centers could face lower utilization than planned. Large projects can also encounter delays involving electricity connections, permitting, financing and community opposition.

PwC estimates that tighter restrictions on international chip trade could reduce cumulative AI infrastructure investment to approximately $25.5 trillion through 2050, about $6 trillion below its central forecast.

AI Infrastructure Could Reshape Real Estate and Energy

For the real estate industry, the AI buildout is creating a new category of high-value infrastructure assets.

Data center developers are competing for locations with available electricity, strong fiber connections and sufficient land. Areas that can provide those resources could attract billions of dollars in investment, while regions with limited power capacity may struggle to participate.

The energy industry faces a similar transformation.

More AI computing means more electricity demand, which could accelerate investment in power generation, transmission and storage. In some markets, data center developers are even exploring dedicated power sources because existing grids cannot provide capacity quickly enough.

The AI boom is therefore becoming much more than a technology story. It is becoming an infrastructure story.

The Next Technology Supercycle

The $31.6 trillion forecast illustrates the scale of what could become a decades-long investment cycle.

AI companies will need more computing power as models become more capable and adoption expands. That demand will require more chips, servers, data centers and electricity, creating a chain of investment extending across multiple industries.

The biggest question is whether the projected growth materializes at the pace expected.

For now, companies and investors are betting heavily that it will.

If AI adoption continues accelerating, the next generation of technology infrastructure could become one of the largest capital investment cycles the global economy has ever seen — reshaping not only computing, but also energy, construction, real estate and industrial development along the way.