Oracle Raises Massive Debt for AI Infrastructure

Reuters reports that Oracle is substantially increasing its financing and data-center investment to support AI infrastructure, and that the company expects to raise nearly $40 billion in 2027 through a combination of debt and equity financing. Reuters also reports fiscal 2026 capital spending reached $55.66 billion, above Oracle's prior target, and Bloomberg places the annual total at $55.7 billion. Reuters states Oracle said fiscal 2027 capital expenditures could be as high as $95 billion, with up to $25 billion of that expected to be repaid by customers. Multiple outlets, including Reuters and Bloomberg, link the spending to large data-center deals with customers such as OpenAI and Meta Platforms. Gizmodo and the Wall Street Journal document Oracle's prior heavy borrowing and earlier debt packages, and Reuters reports shares fell about 8.9% in extended trading after the disclosures.

Major cloud and AI infrastructure buildouts are capital-intensive and frequently financed through a mix of operating cash flow, bank loans, bond issuance, and equity programs. Fortune and Moody's data, cited in Fortune reporting, indicate the hyperscaler group has committed hundreds of billions to data centers, and Bloomberg and Reuters show Oracle's reported capex is now in the tens of billions per year. Reuters quotes Oracle CEO Clay Magouyrk: "Our pace of delivery continues to accelerate with our (fiscal first quarter of 2027) delivery approaching one gigawatt, nearly the same capacity as we've delivered in the previous four quarters combined," reflecting the scale of capacity deployment described in reporting.

Reporting from the Wall Street Journal and FT frames Oracle's financing as part of a wider shift in which hyperscalers are moving risk into debt markets and onto bank balance sheets. The WSJ reports banks struggled to syndicate large loans tied to Oracle-leased sites, constraining traditional lending channels. Fortune and other coverage place this episode in historical context, comparing the current spending sprint to previous infrastructure booms that produced winners and losers. For markets, the combination of record revenues and rising, near-term negative free cash flow tied to capex has heightened investor scrutiny, as reflected in the sharp post-earnings share reaction reported by Reuters and Bloomberg.

Observers should monitor: customer repayment schedules and contractual take-or-pay terms that reporters cite as a partial offset to capex; progress and completion rates on large projects (Reuters reports one project will be more than three-quarters complete within 90 days); bond issuance terms and yield spreads on data-center-backed paper (FT and WSJ coverage flags investor demand and yield pressure); and counterparty exposure among banks that participated in earlier loans, as described by the WSJ. Reporting also flags concentration risk around a few very large customers, notably OpenAI and Meta, which will affect how lenders and bond investors assess credit risk.

The coverage collectively documents a major corporate financing and buildout episode that shifts material risk into public and private debt markets. Practitioners tracking infrastructure costs, total cost of ownership for cloud providers, and financing conditions for large-scale GPU and data-center projects will want to follow reported customer repayment flows, bond-market appetite for AI-linked issuance, and completion milestones that reporters identify as the primary operational offsets to the large capital outlays.