CNBC reported on 29 October 2026 that Microsoft had announced pricing for USD 25 billion in bonds, with the offering expected to close on 4 November 2026. The deal gives Microsoft another financing route as it targets USD 170 billion in 2026 capital expenditures for infrastructure. The proceeds are available for broad corporate purposes rather than earmarked solely for AI capacity.
Debt alongside cash flow and leasing
Microsoft is using the bond market while it continues to report positive free cash flow. On 28 October 2026, finance chief Amy Hood told analysts that the company would use additional funding sources when conditions were attractive. She said leasing was already one financing vehicle, and that Microsoft had access to debt markets because of its ratings.
S&P gives Microsoft a AAA credit rating. That rating matters for the terms on which the company can borrow, particularly while central banks have raised interest rates to fight inflation and investors have become more cautious about large debt balances.
The spending context is Microsoft’s infrastructure expansion for AI demand. Management is targeting USD 170 billion in capital expenditures in 2026, with demand coming from Anthropic, OpenAI and other clients.
Broad use of proceeds
Microsoft said the net proceeds may be used for working capital, capital expenditures, stock repurchases, acquisitions and repayment of existing debt. That wording gives management discretion after the offering closes. It also means the bond sale should not be read as a direct funding line for one named datacentre programme.
For Azure customers and partners, the immediate technical effect is indirect. Microsoft has not attached new Azure regions, service commitments or capacity dates to the bond sale. The financing gives context to the scale of infrastructure planning behind AI workloads, but it does not change a tenant roadmap by itself.
Maturities through 2066
The offering consists of senior unsecured notes split into ten portions. Two USD 1 billion portions carry floating rates, with maturities on 3 November 2028 and 3 November 2029.
The fixed-rate notes mature across a longer period:
- USD 2 billion at 4.85%, due 3 November 2028.
- USD 3.25 billion at 5.00%, due 3 November 2029.
- USD 3.25 billion at 5.20%, due 3 November 2031.
- USD 2.5 billion at 5.35%, due 3 November 2033.
- USD 4 billion at 5.55%, due 3 November 2036.
- USD 2 billion at 6.15%, due 3 November 2046.
- USD 4 billion at 6.25%, due 3 November 2056.
- USD 2 billion at 6.35%, due 3 November 2066.
The largest fixed-rate portions are the USD 4 billion notes due in 2036 and 2056. The final fixed-rate maturity is in 2066 and carries the highest listed fixed rate in the offering.
Cloud sector context
Microsoft is not the only cloud or AI infrastructure company using debt markets. Amazon and Alphabet have also used bonds to support infrastructure projects this year, and CoreWeave has relied on debt as it expands its datacentre footprint. Oracle also unveiled plans to raise USD 10 billion in new debt and said it expected to return to positive free cash flow in the second half of its 2028 fiscal year.
The free cash flow comparison is part of the market context. Microsoft reported positive free cash flow on 28 October 2026, while Alphabet had negative free cash flow for the second quarter in a row. Meta said last week that it had negative free cash flow, and Amazon reported negative free cash flow on 29 October 2026.
How Microsoft funds datacentre capacity for AI workloads, and how those choices affect Azure capacity, cost and platform planning, are questions ECS's Infrastructure topic treats alongside Azure architecture, hybrid cloud, cost and platform engineering. Sessions and workshops give attendees room to compare capacity planning, cloud economics and platform design with practitioner experience from Azure estates. Bring tenant and platform assumptions to discuss before committing to larger Azure or hybrid-cloud plans.
Photo: Coolcaesar, CC BY-SA 4.0, via Wikimedia Commons, cropped
This article was written by the ECS editorial team with the assistance of Amal Agent, run.events' AI assistant, which supported the research, drafting and proofreading.