Demolishing baseless Chicken Little paranoia while simultaneously constructing the world-changing data-center infrastructure of the AI Economy, Oracle posted blowout Q3 numbers yesterday proving to even the most irrational cynics that Oracle is a powerhouse lead of the AI Revolution.
Let’s quickly review a handful of the sky-is-falling forecasts of doom pumped out by self-proclaimed “experts” and punctured by Oracle’s Q3 results:
BS #1: The AI Data-Center buildout is an overcapacity bubble!! Meanwhile, back on Planet Earth, Oracle’s AI Infrastructure business grew by 243% (not a misprint), and CEO Clay Magouyrk said that in spite of Oracle building out new capacity at an astonishing rate, customer demand will exceed supply for some time to come. Not to put too fine a point on it, but overall cloud-infrastructure revenue was up 44% to $4.9 billion. Need more proof? RPO grew by $29 billion in the quarter — exclusively for new AI-infrastructure contracts — climbing an astonishing 325% to $553 billion.
BS #2: Oracle is taking on crushing debt to pay for its AI data centers!! The company has been very clear about its need to raise capital to fund its massive buildout of AI Data Centers. And on this week’s earnings call, principal financial officer Doug Kehring described how Oracle’s initial efforts in that regard have been received:
“In February, we announced our intent to raise up to $50 billion in debt and equity financing, along with a statement that we do not expect to issue any additional bonds beyond this amount in calendar year 2026. Within days of the announcement, we raised $30 billion through a combination of investment-grade bonds and mandatory convertible preferred stock, with a record order book that was substantially oversubscribed.”
Now if this is all a lot of funny-money chasing a phantom market, and if Oracle’s going to the poorhouse because it chooses to leverage debt markets to spend more than $1 billion per week in CapEx to fund those data centers, then why in the wide world of sports would investors “substantially oversubscribe” that borrowing?
As I’ve said many times, Oracle never follows the playbooks used by others — so perhaps its financial model should be viewed as an innovative approach to an unprecedented challenge/opportunity, rather than as a reason for those short on guts and vision to panic and screech about how the sky is falling?
AI Agent & Copilot Summit is an AI-first event to define opportunities, impact, and outcomes with Microsoft Copilot and agents. Building on its 2025 success, the 2026 event takes place March 17-19 in San Diego. Get more details.
BS #3: Oracle was late to the cloud and late to AI and doesn’t even make its own models!! Against a full set of bone-headed and backward-focused caterwaulings, this one manages to stand out as the most absurd. Those caught up in advancing this particular hysteria fail to recognize — or, they recognize but fail to comprehend — that for the past two years, Oracle has been proclaiming loudly that its cloud and AI solutions will be open to all models because that’s what customers want! And as for all the “late to” nonsense, the Q3 numbers mentioned above and the many more to be found here serve to stuff a stinky sock in that particular pie-hole.
BS #4: Oracle’s new CEOs are in over their heads!! The ultimate proof is the quality of the company’s performance, and Oracle met or exceeded the high end of expectations in every category, including profits. As a result, after the earnings call during which CEOs Mike Sicilia and Clay Magouyrk delivered 95% of the Oracle commentary, the company’s market cap has soared by almost $50 billion.
One other bit of evidence for giving very high marks to Sicilia and Magouyrk, who’ve been in their roles for about six months: during the earnings call, Oracle said its total (not just cloud) fiscal-2026 revenue will hit $67 billion, and that for 2027 that number will jump 34% to $90 billion. To me, that’s pretty strong evidence that Oracle founder and chairman Larry Ellison has full faith and confidence in Magouyrk and Sicilia.
Now, about the whiners: look, I get it — some people see the glass as half-full, and others believe a giant monster is sucking up all the liquid everywhere before devouring those others slowly and painfully. But there comes a time when caution, prudence, moderation, and even healthy skepticism have to give way to decisiveness and bold vision, action, and execution.
And on each and every one of those fronts, Oracle is performing brilliantly. And as I’ll describe over the next week or two, it has built a high-performance machine to build out these massive data centers quickly and productively such that very shortly after they go into operation, they’re no longer cost centers but profit centers.
Final Thought
In closing, here are some key Oracle Q3 numbers — and after seeing them, I’m fairly proud of the predictions I made earlier this week for Oracle’s Q3 performance:
- RPO +325% to $553 billion (bigger than AWS and Google Cloud backlogs combined)
- Cloud revenue +44% to $8.9 billion
- Multicloud Database up 531%
- Cloud Database up 35%
- GAAP earnings per share up 24% to $1.27 (so not just hypergrowth, but increasingly profitable hypergrowth).
Nicely done, Oracle!
Ask Cloud Wars AI Agent about this analysis
