AI - Fueled Equipment Spending Surges as American Businesses Bet Big on Data Centers While Housing Falters

webpronews.com · Feb 23, 2026 · Collected from GDELT

Full Article

American businesses are pouring money into equipment at a pace that has surprised economists, driven largely by an insatiable appetite for artificial intelligence infrastructure. The surge in capital expenditure on data centers, servers, and related hardware is more than compensating for persistent weakness in residential construction, painting a complex but ultimately optimistic picture of the U.S. economy heading into the second half of 2025. The latest data from the Commerce Department shows that business spending on equipment rose at an annualized rate that exceeded Wall Street expectations, with technology-related investments leading the charge. According to MSN News, the growth in AI-related capital expenditure has been substantial enough to offset declining investment in housing, which continues to struggle under the weight of elevated mortgage rates and sluggish demand in many regional markets. Data Centers Drive an Equipment Spending Boom The numbers tell a striking story. Technology giants including Microsoft, Amazon, Alphabet, and Meta have collectively committed hundreds of billions of dollars to AI infrastructure buildouts over the next several years. These commitments are now translating into real orders for servers, networking equipment, cooling systems, and the specialized chips that power large language models and other AI applications. The ripple effects extend well beyond Silicon Valley, reaching into manufacturing hubs, construction firms, and electrical utilities across the country. According to recent reporting, nonresidential fixed investment — the broad category that captures business spending on structures, equipment, and intellectual property — has been one of the more resilient components of GDP in recent quarters. Within that category, spending on information processing equipment and computers has been particularly strong, reflecting the corporate rush to build out AI capabilities before competitors do. Analysts at major investment banks have noted that this wave of spending bears some resemblance to the late-1990s technology boom, though with important structural differences that may make it more sustainable. Housing Remains the Economy’s Soft Spot While equipment spending surges, the residential sector continues to lag. Mortgage rates, though slightly below their 2023 peaks, remain in the range of 6.5% to 7%, a level that has effectively frozen many would-be homebuyers out of the market. Existing home sales have been stuck near three-decade lows for months, and new home construction, while somewhat more resilient thanks to builder incentives, has not been strong enough to move the needle on overall investment figures. The divergence between business equipment spending and residential investment highlights a broader theme in the current economy: growth is being driven disproportionately by corporate capital allocation decisions rather than consumer housing activity. This is unusual by historical standards. In most post-recession recoveries, housing has been a leading indicator and a major contributor to GDP growth. The current cycle, shaped by the Federal Reserve’s aggressive rate-hiking campaign and the structural shortage of housing inventory, has inverted that pattern. As MSN News reported, the AI growth story has been powerful enough to paper over the housing weakness in aggregate economic data. The Scale of AI Capital Expenditure Is Unprecedented To appreciate the magnitude of what is happening, consider the scale of announced AI-related capital spending. Microsoft alone has indicated it plans to spend roughly $80 billion on data center construction in fiscal year 2025. Amazon Web Services has signaled comparable levels of investment. Meta Platforms has raised its capital expenditure guidance multiple times over the past year, with its latest projections suggesting spending of $60 billion to $65 billion in 2025, much of it directed toward AI infrastructure. Alphabet has similarly ramped up its spending plans. These figures are not merely aspirational. They are showing up in hard economic data. Orders for nondefense capital goods excluding aircraft — a closely watched proxy for business investment — have been trending upward. Manufacturers of electrical equipment, industrial cooling systems, and backup power generators are reporting strong order books. The construction industry, while struggling on the residential side, is seeing brisk activity in commercial and industrial projects tied to data center development. States including Virginia, Texas, Ohio, and Georgia have become hotbeds of data center construction, attracting billions in investment and creating thousands of jobs. Utility Companies and Power Grids Face New Pressures The AI infrastructure boom is also reshaping the energy sector. Data centers are extraordinarily power-hungry facilities, and the rapid pace of construction is straining electrical grids in some regions. Utility companies are scrambling to add generation capacity, with some turning to natural gas, others exploring nuclear options, and still others signing long-term power purchase agreements for renewable energy. The energy demands of AI have become a significant factor in utility planning, with some estimates suggesting that data center electricity consumption in the United States could double or even triple by the end of the decade. This dynamic has created investment opportunities across the energy value chain. Companies that manufacture transformers, switchgear, and other grid components are seeing demand that far outstrips supply, leading to extended lead times and rising prices. Natural gas producers and pipeline operators are benefiting from increased demand for gas-fired power generation. Even the nuclear industry, long considered moribund in the United States, is experiencing renewed interest as tech companies seek carbon-free baseload power for their facilities. Several major technology firms have signed agreements to purchase power from existing nuclear plants or to support the development of small modular reactors. Economists Debate Whether the Boom Is Sustainable Not everyone is convinced that the current pace of AI-related investment can be maintained. Some economists and market analysts have raised concerns about the possibility of overbuilding, drawing parallels to the fiber-optic cable boom of the late 1990s, which ended in a painful bust when demand failed to materialize as quickly as expected. The argument is that while AI technology is genuinely transformative, the revenue streams needed to justify hundreds of billions of dollars in infrastructure spending have not yet fully materialized for many companies. Others counter that the comparison to the dot-com era is flawed. Unlike the speculative buildout of the late 1990s, today’s AI investments are being made by companies with enormous cash flows and proven business models. Microsoft, Amazon, and Alphabet are not startups burning through venture capital; they are among the most profitable companies in history, with the financial capacity to sustain elevated spending levels for years. Moreover, the demand signals from enterprise customers adopting AI tools appear strong and broad-based, spanning industries from healthcare and finance to manufacturing and logistics. Trade Policy Adds a Layer of Uncertainty Hanging over all of this is the question of trade policy. The Trump administration’s tariff actions have introduced uncertainty into corporate planning, particularly for companies that rely on imported components for their equipment and infrastructure projects. Semiconductor manufacturing equipment, server components, and specialized materials used in data center construction are all subject to potential tariff impacts. While some companies have front-loaded purchases to get ahead of tariff increases, others have adopted a wait-and-see approach that could dampen investment if trade tensions escalate further. The tariff situation is particularly relevant for the technology sector, given the global nature of semiconductor and hardware supply chains. Many of the chips powering AI workloads are designed in the United States but manufactured in Taiwan, South Korea, or other Asian countries. Tariffs on these components could raise the cost of AI infrastructure buildouts, potentially slowing the pace of investment. On the other hand, the administration’s push to reshore semiconductor manufacturing — supported by the CHIPS Act — could eventually reduce dependence on foreign suppliers, though the timeline for new domestic fabrication plants stretches well into the late 2020s. What the Spending Shift Means for the Broader Economy The current configuration of the U.S. economy — strong business equipment spending coupled with weak housing — has implications that extend beyond GDP arithmetic. For workers, it means that job creation is concentrated in technology, construction (particularly commercial and industrial), and energy sectors rather than in residential building trades and real estate services. For policymakers at the Federal Reserve, it complicates the calculus around interest rate decisions, since the economy is neither uniformly strong nor uniformly weak. For investors, the AI-driven equipment spending boom has been a significant driver of stock market performance, particularly for companies in the semiconductor, cloud computing, and data center infrastructure spaces. Nvidia, the dominant supplier of AI training chips, has seen its market capitalization soar past $3 trillion, making it one of the most valuable companies in the world. Other beneficiaries include companies like Vertiv Holdings, which makes cooling and power management systems for data centers, and Eaton Corporation, which produces electrical components used in data center construction. The question facing the U.S. economy in the months ahead is whether the AI investment wave can continue to compensate for weakness in oth