Oracle’s September 10, 2026 earnings release brings the AI infrastructure story into a new reporting period. First-quarter fiscal 2027 revenue reached USD 19.3 billion, up 30% from a year earlier, while cloud infrastructure revenue increased 121% to USD 7.4 billion. Those are reported results, rather than targets for capacity that has yet to arrive.
The release also shows why the earnings headline alone is an incomplete guide to ORCL event risk. Free cash flow remained negative, Oracle completed a substantial common-stock funding program, and its next-quarter earnings growth comparison excludes a large prior-year gain. Each point changes how the release should be interpreted without establishing what the shares or options must do next.
Growth came with a funding requirement
Total cloud revenue, combining infrastructure and applications, rose 62% to USD 11.6 billion. Cloud applications revenue increased 10% to USD 4.2 billion. The infrastructure growth rate therefore should not be applied to the whole cloud business, much less to all of Oracle. Software revenue fell 3% to USD 5.5 billion as customers continued moving toward the cloud, according to the company.
Oracle reported diluted earnings per share of USD 1.56 under GAAP and USD 1.92 on its non-GAAP basis. GAAP operating income was USD 6.7 billion, compared with USD 8.2 billion after the company’s non-GAAP adjustments. These are different accounting measures; choosing the larger number without identifying the basis obscures the comparison.
Operating cash flow reached USD 23 billion, but free cash flow was negative USD 5 billion for the quarter. Positive cash generation from operations can coexist with negative free cash flow when investment requirements are large. The release attributes the investment to supporting the infrastructure business. It does not follow that the operating cash inflow was available entirely for distribution to shareholders.
Oracle also completed the sale of USD 20 billion of common stock, before commissions, through its previously disclosed at-the-market equity program during the quarter. That is an executed financing event. It should not be described as a fresh announcement of an additional USD 20 billion raise or confused with customer revenue.
Backlog and installed capacity answer different questions
Remaining performance obligations reached USD 664 billion, up USD 209 billion year over year. Oracle said it booked more than USD 30 billion of additional AI cloud contracts in Q1 and that the structure of those contracts had no incremental impact on its capital-raising plans.
That last statement is narrower than saying the business no longer needs external funding. It addresses the incremental impact of those new contracts on existing plans. The completed equity program and negative quarterly free cash flow remain part of the same release.
Oracle reported delivering 850 megawatts of additional data-center capacity and more than 300,000 GPUs to AI cloud customers since the end of Q4. Those operational figures help explain the revenue discussion, but neither is interchangeable with recognized revenue or cash profit. Likewise, RPO is a measure of contracted future obligations, not cash already collected or earnings already earned. Delivery schedules, costs and recognition over time still matter.
The next-quarter EPS comparison needs its footnote
For Q2 FY2027, Oracle forecast total revenue growth of 30% to 34% in both U.S. dollars and constant currency. It forecast cloud revenue growth of 65% to 71% in U.S. dollars and 64% to 70% in constant currency. These ranges are management’s outlook, not a second quarter of completed results.
The non-GAAP EPS target is USD 1.85 to USD 1.93 in U.S. dollars. Oracle describes that as growth of 21% to 25% when excluding a one-time investment gain from Q2 FY2026. The footnote explains that the earlier period included a gain from selling Oracle’s interest in Ampere.
Including that gain in the prior-year comparison, Oracle instead expects U.S.-dollar non-GAAP EPS to decline 14% to 18%. The positive and negative growth descriptions can both be accurate because their comparison bases differ. A headline that reports only one without the adjustment can give a misleading impression of the outlook.
For the full fiscal year, Oracle now expects at least USD 90 billion in revenue and non-GAAP EPS of USD 8.10. These forecasts leave execution uncertainty beyond the just-reported quarter. They do not supply a price target or demonstrate how much of that outlook investors had already anticipated.
Why It Matters For Options Traders
An earnings release resolves some uncertainty about the completed quarter while creating a new information set for the remaining life of an option. Here, the relevant distinction is between reported cloud acceleration and the financing, delivery and earnings-comparison questions that remain. Those questions can matter across different expiration dates, even after the scheduled release has arrived.
An option’s premium reflects the underlying price, strike, remaining time and implied volatility, among other inputs. The options Greeks guide explains these sensitivities. A decline in event-related uncertainty can reduce time value, but it does not guarantee that every option loses value: a sufficiently large underlying-price change can work in the other direction. Nor does a strong revenue growth rate establish the direction of the share-price response.
This article does not report a measured post-release ORCL move, a live option chain or a verified pre-earnings implied-move estimate. Comparing a subsequent stock move with an implied range requires timestamped observations from before the release, a defined expiration and a consistent calculation. Using option prices observed after the news as the supposed pre-event benchmark would misstate the comparison.
For a contract close to expiration, the release also does not remove exercise, assignment or liquidity considerations. A quoted midpoint is not proof of an executable fill. Changes in spreads and available size can affect realized outcomes separately from the accuracy of a company’s earnings forecast.
Common misunderstandings and caveats
Triple-digit infrastructure growth does not mean every Oracle business grew at that rate. A large RPO balance is not current-period revenue. Completing a stock sale is financing, rather than operating performance. And Q2 EPS growth excluding the Ampere gain is not the same comparison as growth including it.
These distinctions support a more careful reading of the event; they are not a recommendation to buy or sell ORCL shares or any option strategy. Company forecasts may change, and an option can lose its entire premium. Short options can involve substantial losses and assignment obligations.
This is not financial advice. Options trading involves risk and is not suitable for all investors.
Sources
Oracle, Q1 FY2027 earnings release, September 10, 2026, SEC Exhibit 99.1:
https://www.sec.gov/Archives/edgar/data/1341439/000119312526387905/orcl-ex99_1.htm
Oracle, Form 8-K, September 10, 2026:
https://www.sec.gov/Archives/edgar/data/1341439/000119312526387905/orcl-20260910.htm
Options Industry Council, Options Pricing:
https://www.optionseducation.org/optionsoverview/options-pricing
FINRA, Options:
https://www.finra.org/investors/investing/investment-products/options
