TL;DR
Get smart everyday buys delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
Amazon, Microsoft, Alphabet and Meta spent about $202 billion on property and equipment from July through September 2026, more than twice the year-earlier total, according to company filings compiled by Thorsten Meyer AI. The companies are funding the build-out with a mix of cash, debt, leases, partnerships and equity, while the scale and duration of some commitments leave open who would bear losses if AI revenue lags.
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 compilation of their filings by Thorsten Meyer AI. The surge is being financed through a combination of company cash, borrowing, leases, partnerships and share sales, spreading the costs of the AI data center build-out across a wider range of investors and counterparties.
The quarterly total is a sum of the companies’ reported property and equipment purchases, which include data centers, servers and chips but are not all exclusively for AI. Amazon reported $77.3 billion in spending, Alphabet $51.4 billion, Microsoft $39.4 billion and Meta $33.9 billion. Amazon also builds warehouses; the company said its increase primarily reflected AI investments. The figures and comparisons come from company filings and results cited by Thorsten Meyer AI.
Several funding announcements show how the companies are covering the bill. Meta sold $25 billion in common stock and $12.5 billion in preferred shares during the week covered by the report. Microsoft issued $25 billion in bonds, including debt with a maturity as far out as 2066. Amazon disclosed a loan facility of up to $80 billion secured by servers it plans to buy. The report also calculates that Amazon, Alphabet and Meta raised about $225 billion in long-term debt during the first nine months of 2026.
Cash flow is not covering all current investment at three of the four companies, based on the report’s calculations and company disclosures. Amazon’s trailing 12-month free cash flow was negative $42.4 billion; Alphabet reported negative $9.1 billion and Meta negative $2.1 billion for the quarter. Microsoft remained positive. These measures reflect different reporting periods, and the spending figures include property and equipment beyond AI infrastructure.
Capital, Credit & Compute · Q3 2026
Who’s Funding the AI Data Center Boom?
A guide to the cash, bonds, leases and partnerships behind a $202 billion quarter—and the long commitments that spread the bill across investors and counterparties.
01 / The spending surge
Where the quarter went
The reported total covers property and equipment, including data centers, servers and chips. It is not exclusively AI spending; Amazon also builds warehouses.
Q3 2026 property & equipment purchases · USD billions
Company filing figures compiled by Thorsten Meyer AI. Categories and reporting periods may differ across companies.
Cash flow snapshot
Three below zeroReported free cash flow was negative $42.4B at Amazon, negative $9.1B at Alphabet and negative $2.1B at Meta. Microsoft remained positive.
How to read the total
Big, but not all AIThese purchases include equipment and property for other operations too. Amazon said its increase primarily reflected AI investments.
02 / The funding mix
Five ways to cover the bill
Large technology companies are using cash alongside financing tools that bring lenders, landlords, partners and shareholders into the build-out.
Operating cash
Cash generated by advertising, software and online commerce still helps fund infrastructure investment.
Bonds & loans
Microsoft issued $25B in bonds, with maturities as far out as 2066. Amazon disclosed a server-backed facility of up to $80B.
Leases
Landlords and project owners can finance facilities that technology companies occupy and pay for over time.
Partnerships
Strategic partners or separate facility-owning entities can share the up-front investment and exposure.
Equity sales
Meta sold $25B in common stock and $12.5B in preferred shares during the week covered by the report.
Meta said it intended to finance AI investment through a balanced mix of operating cash flow, investment-grade debt, strategic partnerships and equity capital.
03 / Who carries the exposure?
Long commitments meet short-lived gear
A lease or power contract can last beyond the useful life of the equipment it supports. If demand grows more slowly than expected, the effect depends on contract terms.
Combined future payments on leases not yet started, mostly for data centers, as of September 30, 2026. This is a multi-year commitment total, not a bill due all at once.
Typical depreciation period cited in the report.
Some agreements commit buyers to electricity supply for many years.
Property agreements may continue well beyond the life of individual servers.
Debt can extend across decades, including Microsoft bonds maturing as late as 2066.
04 / Plans & pressure points
What to watch next
Company outlooks point to continued investment, while disclosures on cash, borrowing and lease obligations will show how the funding mix evolves.
About $220B
Expected cash capital spending for 2026. The company said the increase primarily reflects AI investment.
$195B–$205B
Spending guidance remained in this range, with a significant increase expected in 2027.
Higher-cost debt
CoreWeave senior notes carry interest rates of 8.5%–9.75%. This comparison alone does not establish default risk or eventual returns.
The open question
Filings and financing announcements show investment and funding arrangements, but do not establish how quickly AI services will generate revenue or whether returns will cover the costs. Any effects on lenders, building owners, pension funds or electricity customers depend on specific contracts and future demand.
05 / Trace the funding
From capital to consequence
The same build-out can connect company budgets to financial markets, property owners and power suppliers.
The figures describe commitments and spending; they do not predict whether losses will occur or how investments will perform.
06 / Key questions
Quick answers
How much did the four companies spend?
About $202B on property and equipment from July through September 2026, according to figures compiled from company filings. The total includes some spending outside AI.
How are data centers being funded?
Through operating cash, borrowing, leases, partnerships or facility-owning entities, and equity sales. Companies can combine several channels.
What does the $1.02T figure mean?
It is an estimate of future payments on leases not yet started by September 30, mostly for data centers. It is not due immediately.
Does the spending prove AI data centers will be profitable?
No. The cited filings do not establish future AI revenue or returns. Whether revenue will cover investment costs remains unclear.
Who Bears the Financing Risk
The scale of investment matters because the funding methods shift costs and exposure beyond the companies’ immediate operating budgets. A bondholder lends money and expects repayment; a landlord or project company may own a data center and depend on a technology tenant; a utility may commit to supplying power for many years. The report puts the four companies’ future payments on leases not yet started, mostly for data centers, at about $1.02 trillion as of September 30, 2026. That is a compiled estimate from their quarterly filings, not a bill due all at once.
Those commitments can outlast the equipment they support. The source says servers are typically depreciated over five to six years, while some power contracts run 15 to 20 years, data center leases can extend up to 20 years and some bonds run 40 years. If demand or AI revenue grows more slowly than planned, investors, building owners and potentially electricity customers could face consequences, depending on contract terms. The filings and figures cited do not establish that such losses will occur.
Products Worth Considering
FLEXISPOT EN1 Electric Standing Desk, 48 x 24 in, Black
Logitech MX Keys S Wireless Keyboard, Graphite
HP Sprocket Photo Booth Instant Color Printer Bundle
As an affiliate, we earn on qualifying purchases.
A Shift Beyond Cash Funding
The report describes a change from an earlier model in which large technology companies largely funded data centers from cash generated by advertising, software and online commerce. In 2026, the pace and scale of construction have brought multiple financing tools into use at the same time. Capital spending rose sharply across the four companies, while borrowing, leases and equity sales added sources of funding.
Meta stated on October 26 that it intended to finance AI investment “through a balanced mix of operating cash flow, investment-grade debt, strategic partnerships, and equity capital.” Amazon said its 2026 cash capital spending was expected to reach about $220 billion and said the increase primarily reflected AI investment. Alphabet kept its 2026 spending guidance at $195 billion to $205 billion and expected a significant increase in 2027, according to the source material.
Long-Term Costs and Returns
The filings and announcements cited show how much the companies spent and describe some financing arrangements, but they do not establish how quickly AI services will generate revenue or whether that revenue will cover the investment. The source’s lease total is an aggregation of future commitments, and the timing and conditions of payments vary by contract. It is also unclear how much of each company’s property and equipment spending directly supports AI rather than other operations.
The report identifies newer GPU cloud providers, including CoreWeave, as a potential area of greater financing pressure and says its senior notes carry interest rates of 8.5% to 9.75%. That comparison does not by itself show the likelihood of default or the eventual return on data center investments. Any effects on lenders, building owners, pension funds or electricity customers depend on specific contracts and future demand.
Spending Plans and Debt Ahead
The next evidence will come from company filings and results showing whether the announced investment plans continue, how much is funded with cash or new financing, and how quickly facilities are built and used. Amazon’s stated 2026 capital spending outlook is about $220 billion, while Alphabet’s guidance is $195 billion to $205 billion for 2026, with a significant increase expected in 2027. Those are company outlooks, not completed spending.
Investors and customers will also be watching future disclosures on lease obligations, borrowing and operating cash flow. Those figures can help clarify how the cost is distributed, but the returns from AI infrastructure and the full effect of long-term power and property agreements remain uncertain.
Key Questions
How much did the four companies spend in Q3 2026?
Amazon, Microsoft, Alphabet and Meta together spent about $202 billion on property and equipment from July through September 2026, according to figures compiled from their filings by Thorsten Meyer AI. The total includes some spending outside AI.
How are technology companies financing data centers?
The source report identifies five main channels: operating cash, debt, leases, partnerships or separate entities that own facilities, and equity sales. Companies may use several channels at once.
What does the $1.02 trillion lease figure represent?
It is the report’s combined estimate of future payments on leases that had not started by September 30, 2026, mostly for data centers. It is a multi-year commitment total, not an amount due immediately.
Does the spending show that AI data centers will be profitable?
No. The filings and financing announcements describe investment and funding, but the source material does not establish future AI revenue or returns. Whether revenue will cover the costs remains unclear.
Source: Thorsten Meyer AI
