Listen to the audio version of this article (generated by AI).
Key Points
- The performance of Oracle’s stock over the past 18 months should be signaling to investors that the market has serious concerns about the company’s financial health.
- Oracle stock still has significant downside risk. It is heavily exposed to OpenAI, which must continue raising capital to survive, given its persistent losses.
- Oracle’s growing leverage – with more debt issuance on the way in 2027 – the increasing cost of its debt, and layoffs are major concerns for investors.
The stock market is flashing a huge warning sign about Oracle (ORCL) right now. After hitting relative peaks twice over the past two years, the stock has since plummeted, despite Oracle’s heavy spending on artificial intelligence (“AI”), the technology that’s being hyped as the new big growth avenue.
Oracle is in a precarious position right now that could lead to massive downside in the stock. Its fate is heavily dependent on at least one other company in a situation it does not fully control, and a poor outcome could cause it to sink another 50% or so.
Oracle is one of the world’s largest tech companies, and it’s spending big on artificial intelligence data centers, mainly in conjunction with massive deals it has made wth OpenAI. While Oracle sometimes gets thrown in with the big AI hyperscalers – Microsoft (MSFT), Amazon (AMZN), Alphabet (GOOGL), and Meta Platforms (META) – it’s substantially smaller and has fewer resources.
Oracle’s market capitalization is currently hovering around $365 billion, compared with the big hyperscalers valued in the trillions. That’s a reflection of the company’s much smaller $17.1 billion profit in its latest fiscal year, compared with Amazon’s $77.7 billion in 2025, for example.
What those figures don’t show is how poorly Oracle’s stock has performed in the recent past. It can’t seem to hold on to a gain for any length of time, as investors worry about Oracle’s exposure to OpenAI.
For example, back in early 2025, Oracle stock was trading at about where it is today, around $125 per share, before it went on a run above $300 into September. Over the next six months, investors pushed the stock back down to below $140 by April. The stock rallied to $250 by early June, only for it to fall again by some 50% since then, to around $125.
That kind of volatile price action suggests investors are deeply worried about Oracle, particularly as the firm is betting so heavily on AI. These kinds of price swings are not what investors typically see during boom times. Money is still flowing freely, and no one is having trouble meeting their obligations for now.
Investors should brace for some sharp possible declines in Oracle’s stock if funding for AI becomes much tighter in the future.
Oracle’s Success Is Tied to OpenAI
The single biggest risk factor for Oracle is OpenAI, which makes up about half of the remaining performance obligation (“RPO”) of $638 billion that sits on Oracle’s books. The RPO is future contract revenue that Oracle expects to receive from its data-center tenants, such as OpenAI.
OpenAI connects a variety of major players in the AI world, including Cerebras Systems (CBRS), CoreWeave (CRWV), and, of course, Oracle. It’s vital to the AI industry that money keeps flowing through OpenAI, since these companies and others rely on it for revenue. Some of these stocks may well go to zero if the AI bubble bursts.
So, OpenAI is a huge factor in whether Oracle gets paid on its investment. Oracle revealed that it expects about 12% of its total RPO, or $76.6 billion, to arrive in fiscal 2027.
But OpenAI is currently unprofitable, and it may never become profitable. OpenAI lost $20.9 billion on an operating basis in 2025 and an estimated further $7 billion in 2026’s first quarter.
More recently, the business situation seems to have worsened for OpenAI and other American AI models. Chinese AI models have taken significant market share from American models, and that ultimately leads to less revenue for OpenAI. While OpenAI was once a clear leader, it ranked only fourth in token volume for June, according to the AI model brokerage platform OpenRouter.
Chinese AI models are winning market share with token costs at 5% to 10% of those of popular American models, while still offering “good enough” performance for lower-impact uses of AI. Now, even Chinese AI models such as market-leading DeepSeek are looking for funding to go public.
With such poor economics, even OpenAI’s own management expects years of losses, meaning OpenAI must raise funding to stay alive and meet its spending commitments, estimated at $1.4 trillion overall by Barron’s.
This means that Oracle’s investment in AI data centers relies heavily on OpenAI securing funding as well. If OpenAI can’t meet its spending commitments from one of the biggest names in AI, Oracle will have a hard time re-leasing data centers at comparable rates, particularly as OpenAI’s inability to raise money likely coincides with a broader, industry-wide downturn.
But Oracle has little control over whether OpenAI can generate the funding that it needs to survive. So, the AI party must keep going for Oracle to get paid. When you start needing things to go right to make a winning investment, you’re stacking the odds against your investment.
Oracle Is Forced to Raise More Funds for the AI Build-Out
To keep its side of the bargain with OpenAI, Oracle needs to continue building AI data centers.
In fiscal 2027 (which began in June), Oracle expects to spend up to $95 billion on capital investment, substantially higher than the $55.7 billion it spent in fiscal 2026. However, Oracle said it expects reimbursements of up to $25 billion from customers such as OpenAI.
But it needs to raise capital to keep building, an estimated $40 billion in 2027, including debt and a slug of $20 billion in equity, which it’s already announced. However, Oracle already has $130 billion in notes payable and other borrowing, as of May, making it still tougher to raise debt.
In fact, the debt market is becoming less receptive to Oracle. In response to Oracle’s rising credit risk, S&P Global Ratings recently downgraded the company’s debt by one notch, to BBB-, one level above “junk” status. A further decline in Oracle’s rating could put significant upward pressure on the interest rates Oracle must pay on future debt, slowing its ability to borrow.
In addition, the cost to insure Oracle’s existing debt against default (via credit default swaps, a type of derivative) over the next five years has soared since mid-2025 and continues to rise. While these insurance costs have also risen for the big four hyperscalers, the increasing cost is much more pronounced for Oracle, given its lower financial strength.
Of course, if Oracle can’t borrow as much, one avenue is to issue more stock. As mentioned, it’s already planning a $20 billion issuance. This move would likely put further pressure on the company’s stock, which has already sunk sharply over the past month.
If the stock continues falling, it makes a stock issuance even less attractive, potentially leading to a reflexive feedback loop. The stock could fall simply because investors keep expecting Oracle to issue more shares at an ever-declining price, diluting their value even further.
Oracle’s Layoffs Hurt Long Term
To help fund its AI ambitions, Oracle has also laid off a sizable chunk of its employees, about 21,000 in the past year, or about 13% of its workforce. The move frees up money paid to those employees, which can be rerouted to the massive costs of AI data centers.
Oracle took a $1.8 billion restructuring charge for the move, which hurts now. However, a key issue is that this kind of short-term cost-cutting often dogs companies over the longer term, as they lose important institutional knowledge and skills that support the core business. So, the AI build-out likely costs much more than what appears on Oracle’s income statement.
Of course, it’s not just Oracle engaged in this headlong pursuit to cut staff in favor of AI investments. Meta announced layoffs of 8,000 in May, even as it awarded huge stock compensation packages to executives. Microsoft offered voluntary buyouts in April before announcing layoffs of 4,800 in July, about 2.1% of its staff.
With the AI bubble inflating, Oracle is sacrificing its long-term positioning for a risky investment in AI today. Oracle’s success relies heavily on OpenAI’s ability to pay its bills, which in turn depends on ongoing infusions of capital.
With Oracle continuing to issue more debt to fund its AI investments, the company is making a highly risky wager on OpenAI’s success – one with many chances to fail.
Regards,
James Royal, PhD
Editor’s Note: Whitney Tilson called the rise of Apple, Amazon, and Netflix… as well as the collapse of dozens of companies that went bankrupt. Now the former $200M hedge fund firm manager is stepping forward with what he calls the most important financial warning of his 30-year career. He’s sharing two free stock recommendations (one to buy, one to sell immediately) along with details of a new proprietary system fueling his predictions. See it all in his free presentation.
