Microsoft Signs Over $130 Billion in New Data Center Leases as Big Tech's AI Infrastructure Tab Passes $2 Trillion

Microsoft Corp. signed more than $130 billion in new data center leases during the past quarter, according to Bloomberg. The figure represents the company's largest quarterly expansion in data center leasing and pushes its total future lease obligations, where payments haven't yet begun, to $329.1 billion as of June 30.

CEO Satya Nadella told analysts the company is on pace to roughly double its data center capacity. The new leases are primarily for data centers and will support AI demand for years, the company stated in its quarterly filing. This single-quarter commitment from Microsoft exceeds what most enterprises spend on infrastructure over decades.

U.S. technology companies lifted their total future data center lease commitments past $850 billion in the first quarter of 2026, a Bloomberg analysis of quarterly filings shows. That marks a 204% jump from the same period a year earlier, representing a $570 billion increase. The figure also reflects a $200 billion increase from the previous quarter alone.

Meta Platforms and Microsoft drove most of the new activity, each committing tens of billions of dollars in additional leases during their most recent quarters. Oracle joined them as a major contributor to the buildout. The Financial Post and New York Post both reported that Meta and Microsoft are leading the pack of tech giants shoveling money into AI data center leases, with the combined commitments across the largest cloud companies now exceeding $850 billion.

Meta Platforms Inc. reported $279 billion in future lease agreements mostly related to AI data centers that are not yet reflected on its balance sheet, according to Bloomberg. Mark Zuckerberg has said his company will spend hundreds of billions of dollars on AI infrastructure by the end of the decade.

Meta's stock has slid roughly 4% over the past year amid investor anxiety about the profitability of its massive AI data center investments. The Wall Street Journal noted that the price to finance the AI data center boom is rising, pointing to Meta as a case study in how capital costs are climbing for these bets.

Bloomberg reported that the four largest players in the data center race have committed nearly $2.4 trillion in spending over the coming years. This figure encompasses not just lease obligations but broader spending commitments, pointing to massive ongoing investment in AI infrastructure that will reshape the industry for years.

Tech-insider noted that the $850 billion lease figure alone, as of Q1 2026, represents a 63% jump from a year earlier and a 31% increase from the previous quarter. The pace of new commitments shows no sign of slowing, with Microsoft's $130 billion quarter and Meta's $279 billion in off-balance-sheet leases arriving in rapid succession.

Microsoft's results show a clearer path to monetization. Xtb reported that while Meta's results sparked market anxiety regarding the profitability of AI investments, Microsoft's leadership is seeing AI investments start to pay off through its cloud business. The company's financial report for the fourth quarter of the 2026 fiscal year demonstrated that the world's largest software company, now arguably more of a cloud company, is translating infrastructure spending into revenue.

Meta CEO Mark Zuckerberg, meanwhile, is exploring a data center business to generate returns beyond advertising. He told Bloomberg in early July that Meta is exploring leasing computing power to other companies. The New York Times reported that Meta is in talks to lease computing power to Anthropic, signaling a potential new revenue stream that would offset the enormous capital outlay.

Microsoft's disclosure that the $130 billion in new leases are obligations where payments haven't yet begun is a telling detail. The company is locking in capacity years in advance, betting that AI demand will continue to accelerate and that securing physical infrastructure now prevents bottlenecks later. Hostingdiscussion noted that the scale of commitment says a lot about how seriously Microsoft is taking the AI infrastructure race right now.

The total volume of future lease obligations reaching $329.1 billion for Microsoft alone, as reported by UNN, means the company has effectively pre-committed to a footprint that rivals the GDP of mid-sized nations. The structure of these agreements, typically long-term leases with staggered payment starts, allows the company to secure capacity without the immediate balance sheet impact of capital expenditure, though the eventual cash outflows are enormous.

The Wall Street Journal's analysis of Meta's situation highlights a tension that applies across the sector: the cost of capital for these projects is rising, and investors are demanding clearer paths to profitability. Meta's stock performance over the past year reflects this skepticism, while Microsoft's ability to show AI revenue growth has kept its investors more patient.

The divergence between the two companies' market reception suggests that infrastructure spending alone doesn't satisfy Wall Street. The companies that can demonstrate AI revenue growth alongside their infrastructure buildout will face less scrutiny. Meta's exploration of a data center business and its reported talks with Anthropic represent an attempt to bridge that gap.