🔍 Read the full analysis: What It Takes To Pay For AI’s Data Center Buildout on ThorstenMeyerAI.com
Get hardware and tech essentials delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
Amazon, Microsoft, Alphabet and Meta spent about $202 billion on property and equipment in July–September 2026, more than twice the year-earlier amount, according to company filings compiled by ThorstenMeyerAI.com. The report says the buildout is being financed through operating cash, debt, leases, property partnerships and share sales, while the scale and duration of future obligations raise questions about who bears the risk if AI revenues lag.
Amazon, Microsoft, Alphabet and Meta spent about $202 billion on property and equipment from July through September 2026, more than twice their roughly $97 billion total a year earlier, according to a calculation based on their cash-flow statements by ThorstenMeyerAI.com. The spending surge is pushing the companies to combine cash with borrowing, leases, property partnerships and equity financing, raising questions about how the cost of the AI buildout—and the risk that expected revenues arrive slowly—will be shared.
The four companies’ reported quarterly property and equipment spending ranged from $33.9 billion at Meta to $77.3 billion at Amazon. Alphabet reported $51.4 billion and Microsoft $39.4 billion in cash capital spending; Microsoft’s figure rises to $45.2 billion when new finance leases are included, according to the report. The spending covers more than AI alone, but company statements cited in the report link much of the increase to data centers, servers and chips. Amazon said its spending increase primarily reflected investment in artificial intelligence.
The report compares those outlays with cash generated by operations during the same quarter. Amazon generated $42.4 billion, Alphabet $42.3 billion, Meta $32.8 billion and Microsoft $58 billion. On that comparison, three companies spent more on property and equipment than their operations generated in the quarter; Microsoft was the exception. These figures are not the same as a full measure of each company’s available cash, and the report’s free-cash-flow figures use different time periods: Amazon’s trailing-12-month free cash flow was negative $42.4 billion, while Alphabet and Meta reported negative quarterly free cash flow of $9.1 billion and $2.1 billion, respectively.
The financing activity extends beyond cash spending. The report says Amazon, Alphabet and Meta raised about $225 billion in long-term debt in the first nine months of 2026. Microsoft sold $25 billion in bonds this week, including debt maturing in 2066. Meta said on October 26 that it planned to fund AI investments through operating cash flow, investment-grade debt, strategic partnerships and equity capital. The report also cites a newly arranged Amazon loan of up to $80 billion, secured by servers it will purchase with the proceeds, and Meta’s sale of $25 billion of common stock and $12.5 billion of preferred shares.
Infographic 01 · One quarter, four companies
The bill doubled. The cash did not.
What Amazon, Microsoft, Alphabet and Meta paid for buildings, servers and chips from July to September 2026, set against the cash their businesses brought in.
AmazonQ3 2025: $35.1B spent
Spent $34.9B more than came in
AlphabetQ3 2025: $24.0B spent
Free cash flow −$9.1B
MicrosoftQ3 2025: $19.4B spent
$18.6B left over
MetaQ3 2025: $18.8B spent
Free cash flow −$2.1B
Cash purchases of property and equipment as reported; not all of it is for AI (Amazon also builds warehouses). Gaps for Amazon and Microsoft are my arithmetic; free cash flow for Alphabet and Meta as the companies report it. Microsoft’s quarter is its fiscal Q1 2027 (July–September 2026).
Sources: Amazon results, 29 Oct · Microsoft 10-Q, 28 Oct · Alphabet results, 28 Oct · Meta results, 21 Oct. Debt raised: cash flow statements, nine months 2026.
Powered by Thorsten Meyer AI
Who Carries the Buildout Costs
The funding mix matters because it determines who bears the cost if AI demand or revenue falls short. Cash spending and corporate bonds leave much of the financing within the technology companies. Leases and partnerships can place ownership of buildings or other assets with separate entities, while lenders and investors provide capital in return for repayment or a financial stake. Those arrangements do not remove the underlying expense; they change when it is paid and which balance sheet initially carries it.
The report estimates that the four companies had signed about $1.02 trillion in future lease payments that had not yet begun as of September 30, based on its sum of their quarterly filings. That is a reported estimate of future commitments, not an immediate bill or a measure of debt due at once. Still, the scale makes long-term obligations central to judging the buildout, alongside quarterly investment figures.
There is also a timing problem. The report says servers are generally written off over five to six years, while some data-center leases run as long as 20 years and power contracts can last 15 to 20 years. Bonds may stretch further: Microsoft’s sale includes a 40-year maturity. If equipment becomes outdated or AI-related income grows more slowly than expected, companies and their financing partners still have to meet obligations extending beyond the useful life of some hardware.
Products Worth Considering
Apple MacBook Pro 16.2-inch with M5 Pro Chip (18-Core CPU, 20-Core GPU), 48GB RAM, 1TB SSD, Nano-Texture Display, Space Black
Apple MacBook Pro 14-inch Laptop with M5 Pro chip, 18-core CPU, 20-core GPU, 24GB Unified Memory, 2TB SSD, Wi‑Fi 7 – Space Black
Apple MacBook Pro with M5 Max 18-core CPU 20-core GPU 14.2-inch Display 48GB Memory 2TB SSD Space Black
As an affiliate, we earn on qualifying purchases.
A Spending Surge Across Four Giants
The figures in the report are drawn from company filings and reported guidance. Amazon spent $77.3 billion on property and equipment in the quarter, up from $35.1 billion a year earlier, and forecast about $220 billion in cash capital spending for 2026. Alphabet reported $51.4 billion, compared with $24 billion a year before; its 2026 guidance remained $195 billion to $205 billion, with the company expecting a significant increase in 2027. Those forecasts are company guidance, not final spending totals.
Microsoft’s cash spending rose from $19.4 billion to $39.4 billion year over year. It expected about $170 billion for calendar 2026. Meta reported $33.9 billion, up from $18.8 billion, and forecast $135 billion to $145 billion in 2026, including finance-lease payments. The report’s overall comparison uses property and equipment cash-flow figures, so the companies’ guidance and accounting presentations are not perfectly interchangeable.
The report describes five funding channels: operating cash, borrowing, leases, partnerships involving property-owning entities, and equity issuance. It also identifies long-term power contracts as another major commitment associated with data-center construction. The companies’ existing businesses—such as advertising, software and online retail—have historically funded infrastructure investment, but the report argues that the present scale is large enough to require several financing methods at once.
“through a balanced mix of operating cash flow, investment-grade debt, strategic partnerships, and equity capital.”
— Meta, in a statement cited by ThorstenMeyerAI.com on October 26
Revenue Versus Long-Term Obligations
The figures establish the scale of reported spending and several financing commitments, but they do not show how quickly AI services will generate enough revenue to cover the infrastructure costs. The source report presents the possibility of slower revenue growth as a risk, not as a confirmed outcome. It also does not establish that the companies will fail to meet debt, lease or power obligations.
The estimated $1.02 trillion in future leases is a combined calculation from four filings; the report does not present it as a single immediately payable balance. The eventual costs, schedules and commercial terms of individual arrangements may differ. Nor does the supplied material quantify how much of the companies’ spending is exclusively for AI rather than other computing, cloud, warehouse or business infrastructure.
The report points to newer GPU-cloud providers such as CoreWeave and notes senior-note interest rates of 8.5% to 9.75% as an example of financing at the sector’s edges. That comparison does not establish the finances of every such provider, or show that the major technology companies face the same borrowing costs. How much risk rests with tech firms, building owners, lenders, investors, pension funds or electricity customers remains difficult to determine from the figures alone.
Upcoming Spending and Filings
The next indicators will be the companies’ reported capital spending, cash generation and financing commitments in subsequent quarters, alongside updates to their 2026 and 2027 investment forecasts. Those disclosures can show whether spending continues at the third-quarter pace and whether operating cash increasingly covers it or companies rely more heavily on debt, leases and equity.
Investors and other readers will also be watching for evidence of returns from AI products and services, as well as information about data-center utilization and long-term power and lease obligations. The source material does not provide a timetable for when the companies expect AI revenue to match infrastructure costs. Until further filings and guidance provide more detail, the balance between investment, returns and obligations remains an open question.
Key Questions
How much did the four companies spend in the third quarter of 2026?
Amazon, Microsoft, Alphabet and Meta spent about $202 billion combined on property and equipment from July through September 2026, according to a sum of their cash-flow statements by ThorstenMeyerAI.com. The comparable total a year earlier was about $97 billion.
What methods are the companies using to pay for data centers?
The report identifies operating cash, debt, leases, property partnerships and share sales. Long-term power contracts also form part of the broader set of infrastructure commitments.
Does the $1.02 trillion lease figure mean the companies owe that amount now?
No. The report describes about $1.02 trillion in future lease payments not yet started as of September 30, based on its sum of four filings. It is a figure for future commitments, not an immediate payment due all at once.
What is the main financial risk described in the report?
The report highlights a possible mismatch between the life of equipment and the length of financing and infrastructure contracts. If AI revenues grow more slowly than expected, companies and their financing partners would still have to meet obligations tied to leases, debt and power. That is a risk scenario, not a reported failure to pay.
Source: ThorstenMeyerAI.com
