Oracle, once a stalwart of enterprise database software, has seen its credit rating slashed to the edge of junk territory as it embarks on a staggering $250 billion data-centre expansion to fuel its AI ambitions. On July 9, S&P Global Ratings downgraded Oracle to BBB-, the lowest rung of investment grade, placing the company one notch above speculative-grade debt. The move sent shockwaves through bond markets and wiped nearly six percent off Oracle’s stock price as investors reassessed the risks tied to its massive capital spending.
The debt burden grows
Oracle now carries $117 billion in outstanding debt, making it the second-largest non-financial corporate issuer in the Bloomberg US Corporate Bond Index, trailing only Amazon. The company's financial profile has deteriorated sharply as it pours money into building out AI-focused data centres. In its fiscal year ending May 31, 2024, Oracle's free cash flow turned deeply negative, burning through approximately $24 billion after capital expenditures. S&P projects that deficit could more than double to $42 billion in the current fiscal year as construction costs accelerate.
The sheer scale of Oracle's spending is unprecedented for a company of its size. Over the past 12 months, Oracle has invested more than $55 billion in data-centre infrastructure, a figure that dwarfs its annual revenue of around $50 billion. To fund this expansion, the company plans to raise an additional $40 billion through a combination of debt and equity this year, including $20 billion in at-the-market stock sales. These moves dilute existing shareholders and increase leverage, further pressuring the company's credit profile.
Bond market signals distress
The bond market has already priced in the possibility of a further downgrade. Oracle's ten-year bonds now yield roughly 6.5%, well above the average for BBB-rated debt and closer to the BB range that defines junk bonds. According to Bloomberg, that gap reflects investors demanding a premium for uncertainty around whether AI revenue will materialise quickly enough to service the debt. George Catrambone, head of fixed income at DWS Americas, noted that the yield spread indicates the market sees more risk than S&P's single-notch downgrade suggests.
Moody’s has also placed a negative outlook on Oracle, signalling that a second ratings agency sees material downside risk. If Moody's follows through with a downgrade, Oracle could lose its investment-grade status altogether, triggering forced selling by institutional investors that are only permitted to hold high-grade bonds. Such a scenario would further raise Oracle’s borrowing costs and potentially accelerate its debt-servicing challenges.
Why is Oracle spending so much?
Oracle's massive expenditure is driven by a bet that AI workloads will require enormous amounts of cloud computing capacity, and that the company's dedicated AI cloud infrastructure will attract long-term contracts from major players. The most prominent customer is OpenAI, which according to S&P represents roughly half of Oracle's $638 billion in remaining performance obligations (RPO) — the measure of contracted future revenue. OpenAI's reliance on Oracle's cloud for training and inference has created a symbiotic but highly concentrated relationship. If OpenAI shifts its workload to another provider or cuts back on spending, Oracle's revenue pipeline could evaporate.
To mitigate financing risk, Oracle has asked customers to pre-pay for computing components. The company reports that prepaid and customer-supplied hardware for large AI contracts now totals $75 billion. That figure provides some cushion, but it also indicates that customers are wary of committing fully to Oracle's capacity without financial guarantees. Whether those pre-payments will be sufficient to prevent a further downgrade hinges on how quickly Oracle's cloud revenue—which grew 93% in the most recent quarter—can catch up with its borrowing.
Comparisons with hyperscalers
Oracle is not alone in loading up on debt for AI infrastructure. The so-called hyperscalers—Amazon, Microsoft, Google, and Meta—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. But Oracle lacks the cash-flow cushion that protects its larger peers. Google, for example, posted about $73 billion in free cash flow last year, while Oracle’s cash generation has collapsed under the weight of its capital spending. Amazon and Microsoft also have diversified revenue streams from e-commerce and software subscriptions that provide buffers. Oracle, by contrast, is still heavily dependent on its legacy database licensing business, which is growing slowly and facing competition from open-source alternatives like PostgreSQL.
Moreover, Oracle's debt load relative to its earnings is disproportionately high. According to S&P, Oracle's debt-to-EBITDA ratio stands at approximately 5.5x, well above the 2-3x typical for investment-grade technology companies. The company's aggressive financial engineering over the past decade—including the $28 billion acquisition of Cerner in 2022—has layered additional debt on top of an already leveraged balance sheet. The AI data-centre buildout has pushed the company into uncharted territory, and analysts warn that a sustained downturn in AI demand could leave Oracle with stranded assets and unserviceable debt.
Historical context and strategic pivot
Founded in 1977, Oracle built its empire on relational database software that became the backbone of corporate IT systems. Under co-founder Larry Ellison’s leadership, the company pursued a strategy of aggressive acquisitions (PeopleSoft, Siebel, Sun Microsystems) and high-margin licensing. However, the shift to cloud computing in the 2010s caught Oracle flat-footed. It was late to the public cloud market, trailing Amazon Web Services, Microsoft Azure, and Google Cloud by several years. The current AI boom has given Oracle a second chance: by focusing specifically on AI infrastructure—GPUs, high-speed networking, and specialised data centres—it can bypass the general-purpose cloud war and carve out a niche as the go-to provider for AI workloads.
Ellison has been vocal about this strategy, calling AI a “once-in-a-lifetime” opportunity. In recent earnings calls, he emphasised that Oracle is building data centres “as fast as we can” and that demand for AI capacity far outstrips supply. But the financial community remains skeptical. The company's capital intensity is now higher than that of any other major tech firm, and the time lag between spending and revenue recognition creates a dangerous gap. Oracle’s cloud division reported a 93% revenue jump last quarter, but that growth comes off a relatively small base. Even at that rate, it will take years for cloud revenue to cover the capital outlay.
Regulatory and macroeconomic risks
Beyond company-specific factors, Oracle faces headwinds from potential regulatory actions. The Biden administration's executive order on AI safety and the European Union's AI Act could impose compliance costs or restrict certain types of AI infrastructure development. Trade tensions between the US and China may also affect Oracle's ability to source semiconductors and other critical components for its data centres. Additionally, rising interest rates have increased the cost of servicing Oracle's floating-rate debt, further eroding free cash flow.
The broader economic environment adds another layer of uncertainty. If a recession crimps corporate IT spending, companies may delay or cancel cloud migration projects, directly hitting Oracle’s revenue. The company's concentrated exposure to OpenAI also creates single-client risk; any disruption in that relationship could have outsized consequences. S&P's rating report explicitly noted that “a material reduction in RPO or a loss of a major AI customer could lead to further downgrades.”
What lies ahead
For now, Oracle is betting everything on AI demand sustaining at levels that justify its leveraged balance sheet. The company has committed to spending $250 billion on data centres over the next several years, and it has little room to pivot if the market turns. Its bond yields are already trading like those of a junk-rated issuer, and a Moody's downgrade would likely formalise that status. Oracle's share price, down about 10% since the S&P downgrade, reflects investor anxiety over the company's ability to execute on its AI bet.
The next few quarters will be critical. If Oracle can show that its cloud revenue growth is accelerating and that free cash flow is improving, it may regain credibility with rating agencies and investors. If not, the company could face a debt spiral that forces it to cut spending, sell assets, or seek a strategic partner. With $75 billion in customer pre-payments, Oracle has a buffer, but it is finite. The company is playing a high-stakes game, and the outcome will shape not only Oracle’s future but also the broader landscape of AI infrastructure provision.