AI Race Fuels Big Tech’s $650 Billion Spending Surge
Four of the largest technology companies in the United States have projected capital expenditures totaling approximately $650 billion by 2026 — an astonishing influx of funds designated for new data centers and the extensive array of equipment required for their operation, such as artificial intelligence chips, networking cables, and backup generators. The expenditures anticipated by Alphabet Inc., Amazon.com Inc., Meta Platforms Inc., and Microsoft Corp., all striving for supremacy in the emerging market for AI tools, represent an unprecedented surge this century. According to data, each company’s estimated outlay for this year would establish a new record for capital spending by any single corporation over the past decade. The quest for a parallel to the soaring spending forecasts — which emerged alongside the four earnings reports in the last fortnight — necessitates a look back to at least the telecommunications bubble of the 1990s, and possibly to the expansion of the US railroad networks in the 19th century, postwar federal investments in interstate highways, or even the relief programs of the New Deal era.
The ever-larger numbers — in total, an estimated 60 per cent increase from a year ago — signify yet another acceleration in the wave of data centre construction occurring globally. The rapid construction of these extensive facilities, housing rows of buzzing servers driven by costly processors, has strained energy resources, heightened concerns over increased costs for other consumers, and led to tensions between developers and communities anxious about competition for power or water. It also raises the risk that construction spending by a limited group of wealthy companies, which already represents an increasing portion of economic activity in the US, will skew broader economic data. According to Gil Luria, the four companies “see the race to provide AI compute as the next winner-take-all or winner-takes-most market. And none of them is willing to lose.” Last week, Meta announced that full-year capital expenditures could increase to as much as $135 billion — a potential rise of approximately 87 percent. On the same day, Microsoft reported a 66 percent increase in second-quarter capital spending, exceeding estimates, and analysts anticipate it will spend nearly $105 billion in capex for the fiscal year ending in June. The news prompted the second-largest single-day drop in market value for any stock.
Alphabet, established in a garage south of San Francisco in 1998, on Wednesday surprised investors by disclosing a capital spending forecast that surpassed not only analyst estimates but also the expenditures of a significant portion of US industry — it intends to invest as much as $185 billion. On Thursday, Amazon announced a planned $200 billion in capital expenditures for 2026, which also resulted in a decline in its shares during extended trading. In contrast, the largest US-based automakers, construction equipment manufacturers, railroads, defense contractors, wireless carriers, parcel delivery outfits, along with Exxon Mobil Corp, Intel Corp., Walmart Inc., and the spun-off progeny of General Electric — a total of 21 companies — are projected to allocate a combined $180 billion in 2026, based on estimates. Each tech giant has charted a somewhat distinct path to recovering their investments, yet their expenditures are grounded in a common belief: that OpenAI’s ChatGPT and competing tools, which can generate text and exhibit aspects of human reasoning, will assume an ever more significant role in the lives of individuals both professionally and personally. Creating the advanced software models that enable this transition is an exceptionally costly endeavor, necessitating the assembly of thousands of chips, each priced in the tens of thousands of dollars. Thus, the substantial expenses. The expenditure is based on the belief that the final outcomes will lead to significantly increased revenue in the future.
The investments are reshaping companies that, only a few years prior, maintained a modest physical presence, even as their digital services reached billions of individuals. For a considerable duration, Meta and Google parent Alphabet regarded their luxurious corporate campuses and office spaces as a substantial part of their tangible assets. The majority of their expenditures were allocated to salaries and stock grants for the engineers and sales personnel employed at the company. No longer. For the first time in six years, Meta allocated more funds to capital projects than to research and development, which primarily covers engineers’ salaries. At the close of the previous year, the parent company of Facebook and Instagram reported ownership of $176 billion in property and equipment, a figure that is approximately five times greater than the total recorded at the end of 2019. As the figures continue to rise, it remains uncertain whether the companies will successfully realize their ambitious goals. As the construction of data centers has intensified, they are now vying for limited resources, including skilled electricians, cement trucks, and Nvidia Corp. chips produced in Taiwan Semiconductor Manufacturing Co. factories. “There are and will be bottlenecks,” Luria stated.
There is also the matter of how they will finance it. Meta and Google, whose profit primarily derives from digital advertising; Amazon, the leading online retailer and cloud-computing provider; and Microsoft, the foremost seller of business software, each hold a dominant position in their respective industries and possess substantial cash reserves. Their readiness to invest substantial portions of that capital into an AI-driven future indicates that both those reserves and the patience of investors will face scrutiny. “You’ve had these cash-generating machines,” stated Tomasz Tunguz. “Now, all of a sudden they need that cash, and they need more of it, so they’re borrowing.” Tunguz states, “they don’t always end well.” However, during the ascent, he remarked, “they are all huge catalysts for the economy.” What is more certain is that investors who had rushed to buy the tech titans’ stocks over the past year have shown greater hesitance in the face of the skyrocketing capital spending across the board, in some cases selling even when their main businesses — from online advertising and web search to ecommerce and productivity software — have held steady and revenue has exceeded estimates. “What’s causing concern among people? Definitely the analyst narrative and the rhetoric” about the pace at which AI will disrupt businesses, said Steve Lucas. “I would not debate the potential of AI,” he stated. “I would absolutely debate the time frame, and I would passionately debate the economics.”
