Oracle has reinvented itself several times throughout its long and storied history. Beginning in the late 1970s, the company established itself as the creator of what would become—and still remains—the world’s most influential relational database management system. From there, Oracle evolved into a leading enterprise applications provider before transforming into a major cloud infrastructure platform company.
The rebranding of the Oracle CloudWorld conference to Oracle AI World 2025 made it clear that Oracle had embarked on yet another major transformation, placing AI at the center of its long-term strategy. It’s a strategy that once again places Oracle in direct competition with Microsoft, AWS, Google, Meta, and Alibaba—all companies with significantly greater financial resources and, in several areas, stronger existing positions in the broader AI ecosystem.
For Oracle customers, this is more than a change in marketing. It could influence Oracle’s product investments, cloud priorities, support structure and offerings, licensing strategy, sales behavior, and software compliance activity for years to come.
An Ambitious, Expensive, and Bold Transformation
Oracle’s ambitions are mammoth. The company is striving to become the leader in cloud databases and applications, as well as construction and operation of both cloud and machine learning data centers. Oracle has been demonstrating strong cloud growth. For FY2026, the company reported total revenue of $67.4 billion, including $34 billion in cloud revenue. Oracle Cloud Infrastructure revenue grew 77% for the year.
However, competing at the highest levels of the AI infrastructure market requires enormous and continuous funding. Oracle must finance data center operations, expansion, and construction, not to mention the political, legal, and social issues that accompany all modern data centers. On a purely technical level, the physical infrastructure components include GPUs, networking equipment, power capacity, land, construction, and long-term operating commitments. All this revenue must be committed prior to any revenue realization.
Oracle’s challenge is not simply generating revenue. It must generate sufficient returns to cover the capital, operating, political, legal, and financing costs associated with building and operating this infrastructure.
Rising Debt and Financial Pressure
As of May 31, 2026, Oracle carried approximately $129.5 billion in total debt, up sharply from the prior year. That figure is nearly twice Oracle’s FY2026 annual revenue and represents the highest debt burden in the company’s history. The successful companies of the six cities of Silicon Valley do not normally have massive debt responsibilities.
Oracle has also indicated that it expects to continue investing heavily in AI and cloud infrastructure. Therefore, its leverage could remain elevated or increase further, depending on the pace of construction, operating cash flow, customer prepayments, debt and equity financing, asset sales, and future capital requirements.
The company is effectively making large investments today based on the expectation that demand for AI computing capacity will remain strong or accelerate over the coming decades. That may prove to be a highly successful strategy. However, it leaves Oracle more exposed to changes in AI demand, pricing, utilization rates, construction costs, interest rates, technology shifts, and the financial strength of its largest customers.
Beyond the Balance Sheet
Oracle’s reported debt may not capture the full scale of its economic obligations. An April 2026 Wall Street Journal article, “Oracle’s Deluge of AI Debt Pushes Wall Street to the Limit,” examines how portions of Oracle’s data center expansion are being financed through project-finance structures, special-purpose entities, and long-term lease commitments rather than traditional corporate borrowing. When financing becomes so intricate that only the most adept Ivy League minds can understand or follow the loosely connected entities, red flags abound. When the balance sheets become convoluted and the possibility that these arrangements do not appear—at least with clarity that can be understood by your average financier or accountant in a manner that normally reflects conventional debt—there is cause for further consideration. Nevertheless, long-term leases, purchase commitments, and other financial guarantees represent massive financial obligations. That is quite clear.
It’s a cliche, but the metaphor of the proverbial iceberg with only the tip of the ice being visible above the water is apt here. Oracle’s financial exposure may resemble an iceberg: The reported debt is visible above the surface, with additional long-term commitments just below the surface. An iceberg can present a pending catastrophe. Unmanageable debt can be worse.
That does not mean Oracle is concealing debt or violating accounting rules or legal obligations. It means customers and investors should look beyond a single balance-sheet number when evaluating the scale of the company’s AI investment.
One Step Above ‘Junk Status’
On July 9, 2026, S&P (Standard & Poor’s) Global Ratings downgraded Oracle’s long-term credit rating from BBB to BBB-. The new rating leaves Oracle just one notch above speculative grade, commonly known as “junk status,” which has zero positive connotations. Oracle remains an investment-grade company. However, the downgrade reflects concerns about its increasing leverage, the enormous capital requirements of its AI infrastructure strategy, and the changing risk profile of its business. A lower credit rating can increase borrowing costs and reduce financial flexibility. Another downgrade could limit the number of institutional investors permitted to hold Oracle’s debt.
The downgrade does not suggest that Oracle is in immediate financial distress. It does, however, demonstrate that independent credit analysts view the scale and financing of Oracle’s AI strategy as materially increasing the company’s risk.
Growing Dependence on OpenAI
One of the most significant elements of Oracle’s AI strategy is its relationship with OpenAI. The collaboration was first promoted in the early days of the current presidential administration with the enthusiastic support of the regime. OpenAI has reportedly agreed to purchase approximately $300 billion of computing capacity from Oracle over roughly 5 years. That constitutes one of history’s largest cloud-computing agreements.
The potential revenue is vast, but so is the risk existentially consequential. S&P Global Ratings estimates that OpenAI could account for more than one-third of Oracle’s total revenue by FY2028 and an even larger percentage by FY2030. S&P has also noted that OpenAI’s ability to fulfill its contractual commitments depends on continued AI market growth, access to outside financing, and its ability to remain a leading AI platform. The mysterious nature of pre-IPO OpenAI along with its recent legal battles with Elon Musk and its remaining status as a nonprofit are all aspects of this situation that are causes for concern.
OpenAI is growing rapidly, but it is not profitable, as of this writing. It continues to spend heavily on research, computing capacity, infrastructure, and product development. Reports indicate that OpenAI generated approximately $13 billion in revenue during 2025 while recording substantial operating losses.
Oracle is committing extraordinary amounts of capital to serve a customer that remains dependent on continued uncertain and speculative revenue growth and access to financing. This does not mean OpenAI will default or fail to meet its commitments. It does mean that Oracle’s future financial performance is becoming increasingly linked to the success, financing capacity, and long-term competitiveness of a single customer.
Oracle’s AI Bet
Oracle’s AI strategy represents one of the most consequential bets in the company’s history. It’s fair to suggest that this financial gambit exceeds any that have been attempted before in the six cities of Silicon Valley. Its success depends on several assumptions, which are as follows:
- Demand for AI computing capacity will continue growing rapidly.
- Customers will use the capacity Oracle is building.
- AI infrastructure pricing will remain strong enough to generate attractive margins.
- Oracle can complete data centers on time and within budget.
- Oracle can navigate the complex legal, political, and social roadblocks that are rapidly gaining notoriety to build these massive data centers.
- OpenAI and other major customers can meet their long-term contractual obligations.
- Revenue growth will outpace capital expenditures, operating costs, and financing expenses.
Even modest changes in demand, pricing, financing costs, construction schedules, or utilization rates could materially affect the strategy’s financial returns. Neither Oracle nor its CTO and co-founder, Larry Ellison, should ever be underestimated. The company has repeatedly demonstrated an ability to adapt, compete, and monetize its enormous customer base. Nevertheless, the financial stakes are significantly higher than during many of Oracle’s previous transformations.
Workforce Reductions and Customer Service
Oracle has also implemented significant cost-reduction and organizational restructuring initiatives while increasing its AI investments. Industry analysts and media reports have estimated that Oracle could eliminate between 20,000 and 30,000 positions, although Oracle has not publicly confirmed a final companywide figure. Reported reductions have affected Oracle Health (previously, Cerner), cloud, sales, customer success, engineering, and other functions.
Workforce reductions of this scale can affect institutional knowledge, customer relationships, product development, account management, and support responsiveness—all of which have produced reliable, constant, and massive revenue streams for Oracle over many decades. Customers may experience the following difficulties:
- Longer support queues and slower issue resolution
- More frequent changes in account and support personnel
- Reduced access to experienced technical specialists
- Greater reliance on automation and offshore support resources
- Less investment in products that Oracle considers nonstrategic
- Increased pressure to migrate toward Oracle’s preferred cloud and AI offerings
Public discussions on Reddit, Blind, LinkedIn, X (formerly Twitter), and other social platforms contain growing numbers of anecdotal complaints from customers and former employees about response times, support quality, and organizational disruption. Those comments are anecdotal and have no scientific or statistical meaning. Layoffs do not definitively result in a measurable decline in service. They do, however, indicate that some customers and employees perceive a change in the Oracle support experience. Organizations should monitor their own service metrics, escalation times, support-ticket aging, and access to knowledgeable Oracle personnel rather than relying solely on public commentary.
Could Oracle Divest Non-Core Businesses?
Given Oracle’s capital requirements, it would not be surprising if the company eventually evaluated opportunities to monetize non-core assets or business units. Oracle’s current financial position did not develop overnight. In June 2022, it completed its approximately $28.3 billion acquisition of Cerner—now Oracle Health—which remains the largest acquisition in Oracle’s history. Much of that transaction was financed with debt.
The Cerner acquisition was followed by Oracle’s unprecedented investment in AI infrastructure, data centers, and GPUs. Together, those decisions help explain why Oracle’s leverage and long-term financial commitments have changed so dramatically. Whether Oracle ultimately sells Oracle Health, or another business, remains entirely speculative. Companies undertaking major capital-investment programs commonly review their portfolios to determine whether every asset continues to support their strategic priorities.
However, divestitures are complex transactions that frequently take many months to negotiate and complete. They may improve a company’s long-term financial position, but they are generally not a quick source of readily available cash.
Who Buys It Matters
Should Oracle eventually divest a business or product line, the identity of the buyer could be as important to customers as the sale itself. Changes in ownership can lead to new product strategies, licensing policies, support models, pricing structures, and investment priorities. VMware provides a recent example. Following Broadcom’s acquisition, many VMware customers encountered substantial changes in licensing, product packaging, partner relationships, and commercial terms.
Every transaction is different, and a future Oracle divestiture would not necessarily follow the same path. Nevertheless, the VMware example demonstrates how quickly the customer relationship can change after a major acquisition. Organizations that depend on Oracle products should therefore consider ownership-change risk as part of their long-term technology and licensing strategies.
What This Means for Oracle Customers
Oracle’s strategy is bold, and it may ultimately succeed. However, it carries significant financial and operational risk. As Oracle directs more capital toward AI infrastructure, protecting and expanding revenue from its installed software customer base becomes increasingly important. Customers should expect Oracle to remain highly focused on the following:
- Recurring software and cloud revenue
- Java licensing
- Cloud migrations
- Support renewals
- Contract compliance
- Customer retention
- Expansion of existing accounts
- Converting on-premises workloads to Oracle cloud services
Software compliance activity is particularly important to monitor. Oracle has long used formal audits and less-formal license-review processes to identify compliance exposure, influence purchasing decisions, and encourage cloud migrations.
If Oracle faces pressure to accelerate revenue or improve cash flow, it could place even greater emphasis on compliance enforcement and what is often referred to as an Ambush Audit (a trademark jointly owned by LicenseFortress and Beeman & Muchmore LLP), which is an informal compliance inquiry that may begin without an official audit notice. That does not mean every Oracle customer will be audited. It means organizations should not wait for an audit letter before understanding their licensing position.
Customers should review their Oracle contracts, establish accurate deployment records, monitor Java and database usage, evaluate virtualization risks, document entitlements, and prepare an audit-response plan before Oracle initiates a compliance discussion. Oracle is still competing against larger and more deeply funded companies in the overall AI market. Its strategy will not unfold in a straight line, and customers should expect periods of aggressive investment, organizational change, financial pressure, and shifting commercial priorities. As we wrote in our earlier column, “Oracle—The 100-Year Plan”: “Oracle should not be underestimated.” But neither should the risks created by the scale of its AI ambitions.