S&P Global Ratings delivered a stark warning to Oracle on July 9, downgrading the technology giant's credit rating to BBB- — just one notch above junk status — as the company's massive artificial intelligence data-center expansion continues to consume cash faster than revenue can replenish it. The downgrade reflects growing concern that Oracle's $250 billion capital spending plan over the next several years will strain its balance sheet to the breaking point, especially if AI-driven revenue growth fails to materialize as quickly as executives have promised.
The magnitude of Oracle's AI bet
Oracle has become the second-largest non-financial debt issuer in the Bloomberg US Corporate Bond Index, trailing only Amazon, with $117 billion in outstanding bonds. The company's free cash flow for its fiscal year ending May 31 turned deeply negative, burning through nearly $24 billion after accounting for capital expenditures. S&P estimates that figure could widen to $42 billion as Oracle continues building data centers at an unprecedented pace — spending more than $55 billion on these facilities in the last fiscal year alone.
To finance this expansion, Oracle expects to raise another $40 billion this year through a combination of debt issuance and equity sales, including $20 billion in stock sales at market prices. The company has also asked customers to pre-pay for computing components, with prepaid and customer-supplied hardware for large AI contracts now totaling $75 billion. This approach provides some short-term relief but also ties Oracle's fortunes more closely to the volatile AI market.
Wall Street reacts
The bond market is already acting as if the downgrade has further to go. Oracle's ten-year bonds yield roughly six and a half percent, well above the BBB index average and closer to the BB range that marks junk territory, according to Bloomberg data. George Catrambone, head of fixed income at DWS Americas, told Bloomberg the yield gap reflects investors demanding a premium due to uncertainty about whether AI revenue will justify the debt load.
Shares of Oracle fell nearly six percent on the day the downgrade was announced, as investors digested the implications of the ratings action. Moody's has also placed a negative outlook on the company, signaling that a second major ratings agency sees material risk in Oracle's trajectory. A further downgrade by either agency would push the company into junk territory, triggering forced selling by investment-grade-only funds and potentially raising borrowing costs even higher.
The OpenAI dependency
Perhaps the most concerning aspect of Oracle's strategy is its heavy concentration on OpenAI. S&P estimates that roughly half of Oracle's $638 billion in remaining performance obligations — its measure of contracted future revenue — is tied to the AI research firm. While OpenAI's rapid growth has been a boon for Oracle's cloud business, which grew 93 percent last quarter, such dependency exposes the company to significant single-client risk. If OpenAI slows its spending or moves workloads to another cloud provider, Oracle's revenue pipeline could collapse.
Oracle's cloud revenue growth remains impressive, but it is starting from a relatively small base. The company's overall cloud business accounted for roughly $18 billion in revenue last fiscal year, compared to Amazon Web Services' $90 billion and Microsoft Azure's estimated $100 billion. To justify its massive capital expenditures, Oracle needs to maintain triple-digit growth rates for several years — a tall order in an increasingly competitive market where hyperscalers are all investing heavily.
Comparing to hyperscaler cash flows
Oracle is not alone in loading up on debt for AI infrastructure. Hyperscalers collectively plan to spend up to $725 billion on AI this year, and Big Tech's combined AI debt has already hit $350 billion, according to Bloomberg. However, Oracle lacks the cash-flow cushion that protects its peers. Google, for instance, posted about $73 billion in free cash flow last year, giving it ample room to service its debt even if AI investments take longer to pay off. Microsoft and Amazon also generate substantial free cash flow, thanks to diversified businesses that include advertising, e-commerce, and enterprise software.
Oracle's free cash flow, by contrast, has collapsed under the weight of its capital spending. The company reported free cash flow of just $5.8 billion in fiscal 2024, down from $9.2 billion the prior year, and the trend is expected to worsen before it improves. D&A charges are rising rapidly as new data centers come online, further pressuring earnings. While Oracle's management has expressed confidence that AI revenue will eventually outpace capital spending, the timing remains uncertain.
What's at stake
The stakes for Oracle could not be higher. The company is betting everything on AI demand sustaining at levels that justify turning one of the most leveraged balance sheets in technology into something even more leveraged. Chairman and CTO Larry Ellison has been the driving force behind the strategy, personally pushing for the aggressive data-center buildout. CEO Safra Catz has focused on financial engineering, including the $20 billion equity sale announced in March, which will dilute existing shareholders but provide breathing room.
Oracle's leverage ratio, measured as total debt to EBITDA, has climbed to over 5x, well above the 3x threshold that typically signals high risk. The company has been able to maintain access to capital markets thanks to its long history of stable cash flows and strong relationships with bond investors. But each downgrade erodes that trust. If AI revenue fails to ramp up quickly enough, Oracle could find itself trapped in a vicious cycle: higher borrowing costs reduce earnings, prompting further downgrades, which raise borrowing costs even more.
Market context
The broader AI infrastructure spending frenzy has drawn comparisons to the dot-com bubble, albeit with more concrete revenue streams. Companies like Nvidia, AMD, and TSMC have seen enormous demand for chips and manufacturing capacity, but the end customers — cloud providers like Oracle, Amazon, Microsoft, and Google — are taking on massive debt to build out data centers. The fear is that overcapacity could lead to a price war, destroying margins and leaving some players unable to service their debt.
Oracle's position is especially precarious because it lacks the economies of scale of its larger rivals. Its cloud market share remains in the low single digits, and it has historically focused on running legacy enterprise workloads rather than AI training. The pivot to AI required building new, specialized data centers from scratch, which is far more expensive than retrofitting existing ones. Moreover, Oracle's customer base — mostly large enterprises — may be slower to adopt AI than the startups and developers that favor AWS or Azure.
Despite these risks, Oracle's management remains bullish. The company has guided for double-digit revenue growth in the coming years, driven by what it calls "hyper-growth" in its cloud infrastructure business. Executives have pointed to the $75 billion in prepaid and customer-supplied hardware as evidence that demand is real and long-term. But as the S&P downgrade makes clear, the market is not yet convinced that the math adds up.
The bond market's skepticism is reflected in Oracle's credit spreads, which have widened significantly since the start of the year. At current levels, Oracle pays an interest rate about 200 basis points higher than similarly rated companies. Each basis point adds $11.7 million in annual interest expense on the $117 billion of debt, meaning the spread costs Oracle roughly $2.3 billion per year in extra financing costs. If spreads widen further, that number could rise, cutting into already thin margins.
Oracle's next major test will come when it reports fiscal first-quarter results in September. Analysts will be scrutinizing cloud revenue growth, free cash flow trends, and management's commentary on capital spending plans. Any sign that AI demand is softening could trigger a sharp sell-off in both bonds and equity. Conversely, strong results could reassure investors and potentially reverse some of the negative rating actions.
For now, Oracle remains in a race against time: it must generate enough AI revenue to cover its debt service and capital spending before the market loses patience. The S&P downgrade is a clear signal that patience is wearing thin. Whether the company can pull off its bet on AI will be one of the most closely watched stories in technology finance over the next two years.