Big Four IT Giants to Spend Almost $2.4 Trillion on AI Infrastructure

Alphabet, Meta, Microsoft, and Amazon plan to invest nearly $2.4 trillion in AI infrastructure over the next few years, according to Bloomberg. This sum includes commitments for leasing, building construction, energy, and equipment for data center expansion. Companies justify the spending by citing massive demand for AI computing power, with obligations spanning short-term costs and decades-long commitments. Alphabet, which includes Google, disclosed contractual obligations and leases not yet in effect totaling $902 billion, more than nine times higher than a year earlier, including tech equipment, energy, and rental agreements. Meta's future investments amount to nearly $700 billion, about half of which is for data center leases that have not yet started and will be paid over 30 years; Meta's total commitments grew more than eightfold in a year. Investors criticized Meta after it forecast disappointing sales and raised its lower bound for annual capital expenditures to $130 billion from $125 billion. Amazon increased its annual capex forecast to $220 billion from $200 billion, with CEO Andy Jassy saying even that would be insufficient to meet high demand. Alphabet increased its spending forecast by $15 billion to $205 billion, and Microsoft said component price rises would add $25 billion to its costs, with CEO Satya Nadella noting the company opened 31 data centers on five continents in the last quarter, aiming to roughly double its total capacity in two years. From April to June, the four companies' capital expenditures totaled $170 billion, up 72% year-over-year. According to FactSet, Amazon, Google, Meta, and Microsoft are expected to spend $1.5 trillion on data centers and advanced chips this year and next. Companies are increasingly using debt to finance plans; Nikkei Asia estimates five tech giants (Alphabet, Microsoft, Amazon, Meta, and Oracle) have accumulated about $3 trillion in debt, including $1.65 trillion off-balance-sheet. The key question for investors is whether the hundreds of billions spent on AI infrastructure will pay off.