Oracle's Debt Balloons to $108 Billion as AI Spending Soars

Oracle is rapidly accumulating debt to fund capital-intensive AI infrastructure deals—its remaining performance obligations total $523 billion (including $68 billion added in Q2 FY26)—because many contracts require building GPU-filled data centers before revenue is recognized. For the six months to Nov. 30 the company generated $10.2 billion of operating cash flow versus $20.5 billion of capex, producing negative free cash flow of $10.3 billion and pushing total debt to about $108 billion (up from $92.6 billion in May) after an $18 billion bond sale, a dynamic that has pressured the stock. Management says it expects to need less additional financing than some estimates (KeyBanc projected up to $100 billion over four years for the OpenAI deal) and is exploring customer-chip arrangements and third-party funding to preserve its investment-grade rating, but conversion of the AI backlog into revenue and counterparty funding risks (notably OpenAI) are pivotal to the company’s financial outlook.

Analysis

Oracle has amassed $523 billion of remaining performance obligations and added $68 billion to that backlog in Q2 FY26 on AI infrastructure deals that require building GPU-filled data centers before revenue can be recognized, creating a large but front-loaded capital commitment. For the six months ending Nov. 30 the company generated $10.2 billion of operating cash flow while spending $20.5 billion on capital expenditures, producing negative free cash flow of $10.3 billion and reflecting capex that more than tripled year-over-year.

Total reported debt stood at about $108 billion at the end of Q2, up from $92.6 billion in May following an $18 billion bond sale in September, and the company is borrowing to fund its AI buildout even after years of debt-financed buybacks. Analysts at KeyBanc estimated Oracle might need an incremental $100 billion of debt over four years to fund the reported OpenAI-related commitments, a projection management says will likely be substantially less.

Management has cited alternatives — customer-supplied chips, third-party/private-equity funding structures (an example cited was Intel’s Brookfield deal) — and reiterated a commitment to maintain an investment-grade rating, but conversion of the AI backlog into recognized revenue is the pivotal catalyst. The reported $300 billion OpenAI contract and the company’s worsening cash/debt profile have driven investor skepticism and negative sentiment (sentiment_score -0.55; ORCL -0.7), leaving near-term downside risk if funding or counterparty outcomes deteriorate.

Key Decisions for Investors

• Trim or avoid adding to ORCL positions until management provides clear, executable funding plans or evidence of non-recourse/partner financing that limits balance-sheet risk, given negative free cash flow and rising debt
• Use near-term quarters to monitor specific triggers — cadence of backlog conversion into recognized revenue, capex run rate, changes in total debt and any third-party financing announcements — and reweight only after observing sustained progress
• If seeking AI exposure, prefer staged entry or hedged positions (options or position size limits) rather than outright leverage; consider allocating to counterparties that provide capital-light AI services or to firms with clearer funding models
• Watch credit-market signals closely (rating agency commentary, bond yields, covenant changes) as any deterioration in investment-grade status would materially change risk/reward and warrant rapid position reduction