Situation report: Oracle has launched $20 billion in stock and bond sales to pay for its AI infrastructure expansion, The Information reports. The money is split about evenly between new shares and new debt. This is one of the biggest single capital raises tied directly to the AI data center race. For anyone tracking who can afford to keep building AI compute, it’s a clear signal: even the giants now need Wall Street to fund the next phase.
🎯 What Happened
Oracle went to market fast and priced both parts of the deal within a day. Here are the key facts from coverage of the offering:
- Equity: Oracle sold about 86.2 million shares at $116 each. That’s roughly an 11% discount to the prior close, and it’s a steep price for a company of Oracle’s size.
- Debt: The other half came as bonds in several tranches that mature between 2029 and 2046. Reported coupons on the fixed-rate tranches run from about 7% to 8.65%. Reports differ on the exact number of tranches.
- Stated use: “General corporate purposes,” including capital spending, paying down debt, acquisitions, dividends and buybacks.
- Market reaction: The stock fell about 4.5% on the news. It has lost more than $60 billion in market value since Oracle’s October 28 investor day.
🧭 Why the Timing Raised Eyebrows
At that investor day, Oracle said it would cap new capital-markets fundraising at $20 billion through fiscal 2030, not counting debt refinancing. Five days later, it raised the whole amount in one shot.
That’s technically within the promise. But investors heard “we’ll be careful” and then saw the full budget spent in a week. What stands out here is the urgency. A company doesn’t front-load four years of fundraising unless the bills for its buildout are coming due soon.
📈 How We Got Here
This isn’t Oracle’s first trip to the capital markets this year. The tally so far:
- A $25 billion debt deal completed in February
- A $20 billion stock offering around the same time
- A $20 billion at-the-market stock program finished in Q1 of fiscal 2027
- And now another $20 billion in stock and bonds
Two years ago, Oracle was a steady database and enterprise software business with predictable cash flow. Now it’s one of the most aggressive builders of AI cloud capacity, backed by huge compute contracts with frontier labs. Those contracts promise big revenue later. The data centers, chips and power deals cost money now. That gap gets filled with debt and fresh shares.
⚠️ Threat Assessment
This is significant because Oracle is becoming the test case for financing AI infrastructure on borrowed money. Microsoft, Google, Amazon and Meta can pay for much of their capex from operating cash. Oracle can’t, at least not at its current pace.
The risks in plain terms:
- Leverage: Bond coupons of 7% to 8%+ are expensive for an investment-grade tech name. They show lenders want a premium to take on Oracle’s AI exposure.
- Dilution: Selling stock at an 11% discount hurts existing shareholders, and Oracle has done it repeatedly this year.
- Customer concentration: A big chunk of Oracle’s future cloud revenue depends on a handful of AI labs keeping their spending commitments.
- Contagion signal: If Oracle’s cost of capital keeps climbing, it tells you how much patience markets have left for the wider AI buildout.
🛠️ What It Means for Practitioners
If you buy compute or build on AI cloud platforms, this matters more than it might seem:
- Capacity is still coming. Oracle is clearly committed to bringing more GPU capacity online, which should help with availability over the next 12 to 24 months.
- Pricing pressure could shift. Expensive capital has to be paid back somehow. Watch for changes in long-term contract terms and reserved-capacity pricing.
- Vendor diversification matters. Teams relying heavily on one AI cloud provider should keep a backup plan, especially with a provider carrying this much debt.
🔭 Outlook
Oracle now has its war chest. The next few quarters will show whether its AI cloud revenue ramps fast enough to justify the spending and calm the bond market. If it does, Oracle looks like a bold bet that paid off. If it doesn’t, this raise will be remembered as an early warning about the AI infrastructure boom. Full details are available in The Information’s original report.