Cloud, debt and AI promises: the Oracle checklist before earnings

Key takeaways

• Oracle reports Q2 fiscal 2026 earnings on 10 December 2025, with investors focused on AI cloud growth and backlog.

• Q1 results showed 12% revenue growth and a 359% jump in cloud backlog, but also heavy spending and rising debt.

• Long-term investors can watch cloud growth, backlog conversion and balance sheet discipline instead of trying to guess the headline number.

Oracle walks into this week’s Q2 fiscal 2026 earnings with one of the biggest AI backlogs in the market and a more cautious shareholder base than a few months ago. The company reports after the US close on 10 December 2025, with the call set to focus less on last quarter’s profit and more on how fast Oracle can turn promised AI workloads into real cash.

For long-term investors, the question is not whether Oracle beats consensus by a cent. It is whether the numbers and guidance still support the story of a legacy software giant turning itself into a durable AI and cloud infrastructure platform.

The story since last quarter: from AI euphoria to hangover

Oracle’s Q1 fiscal 2026 results in September delivered a rare mix of solid growth and jaw dropping backlog numbers. Revenue rose 12% year on year to 14.9 billion USD, with total cloud revenue up 28% to 7.2 billion USD and cloud infrastructure up 55%.

The real headline, though, was remaining performance obligations (RPO), a measure of contracted future revenue. RPO jumped 359% year on year to 455 billion USD, helped by several multi-billion AI cloud deals, widely reported to include a huge contract with OpenAI.

Initially, the market loved it. Oracle’s share price spiked on the Q1 release as investors tried to price in years of AI demand landing on Oracle’s data centres. Then the mood flipped. Once it became clear how much of that backlog came from a single customer, and how much capital spending would be needed to serve it, the stock dropped about a third from its high.

Since then, scepticism has centred on two themes. First, whether OpenAI and other AI partners can actually deliver the revenue implied by those contracts. Second, how far Oracle’s leverage and interest costs can rise before they start to constrain flexibility for dividends, buybacks or further expansion.

Where the growth is really coming from

Behind the noise, Oracle’s business mix is changing quite fast.

The growth engine today is Oracle Cloud Infrastructure (OCI), the part of the group that sells raw computing power, storage and networking for AI training, inference and traditional workloads. OCI revenue grew 55% in Q1, and management has guided for 77% growth this fiscal year to about 18 billion USD, with even faster growth pencilled in for the following years based largely on already signed contracts.

On top of that sit Oracle’s cloud applications. Fusion and NetSuite enterprise resource planning software, along with healthcare systems from the Cerner acquisition, continue to grow at healthy double-digit rates. These businesses look less exciting than AI data centres, but they add subscription style, recurring revenue that can smooth the cycle if AI spending ever takes a pause.

Legacy software and on premises licences now play a different role. Growth here is slow or flat, but these products still generate sizeable cash that can help fund the build out of new cloud regions and AI capacity. The risk is that higher depreciation and interest costs from that build out squeeze margins before the newer businesses fully scale.

In short, the numbers this week will tell investors how smoothly Oracle is shifting from a high margin, slower growth licence model to a heavier, more capital intensive AI infrastructure model.

What the market is expecting into this print

Wall Street is looking for another strong quarter on the surface. Bloomberg consensus points to Q2 fiscal 2026 revenue around 16.2 billion USD, up roughly 15% year on year, and adjusted earnings per share of about 1.64 USD, growing in the low teens.

That sits above Oracle’s own guidance from September, which implied more modest earnings per share growth for Q2, and leaves some room for disappointment if margins come under pressure from data centre build outs.

Investors are likely to focus on three lines more than the headline earnings per share figure:

1. Cloud infrastructure growth: does OCI growth stay near or above the 50% mark needed to support management’s full year targets?

2. RPO and bookings quality: does total backlog grow again, and do management comments suggest a broader customer base beyond a few mega AI deals?

3. Margins and capital expenditure: how quickly are depreciation and interest costs rising, and what does that imply for free cash flow in the next couple of years?

With the share price still recovering from its autumn correction, “good enough” numbers may not move the needle unless they come with reassuring messages on balance sheet discipline and AI customer diversification.

Risks to watch on the night

The first risk is execution. Oracle must actually build and power the data centres that sit behind its huge backlog. Delays, cost overruns or supply constraints could slow the pace at which RPO turns into recognised revenue.

The second risk is financial. Net debt and interest costs have risen as Oracle funds capacity for AI workloads, while free cash flow has dipped into negative territory in recent quarters. If that continues, rating agencies and income focused shareholders may get more nervous.

The third risk is customer concentration and AI hype. A large portion of Oracle’s backlog appears linked to a small group of AI leaders, including OpenAI, which itself faces rising competition from rivals and questions over long term monetisation. Any sign of contract resizing, delay or renegotiation would be taken seriously by the market.

Early warning signs to scan for in the release and call include slower OCI growth, a flattening of RPO, more cautious comments on capital expenditure, or vague language around major AI customers.

Investor playbook: how to read the earnings

• Think in trends, not quarters. Track whether cloud growth, RPO and margins broadly match Oracle’s multi-year guidance, rather than fixating on a single print.
• Separate “backlog story” from “cash story”. A bigger RPO only helps if it turns into revenue and, eventually, free cash flow that can service debt and reward shareholders.
• Watch balance sheet flexibility. Follow net debt, interest costs and capital expenditure over several quarters to see whether Oracle can grow aggressively without boxing itself in financially.
• Use scenarios. Consider how your long-term view would change if OCI growth stays above 50%, drops to 30%, or accelerates further, instead of anchoring on a single forecast.

Big AI promises need boring delivery

Oracle’s Q2 report is unlikely to settle every debate about its AI future in one evening. It can, however, show whether the company is quietly doing the hard, boring work of turning a spectacular backlog and ambitious guidance into steady, repeatable progress.

Since September, the market has given Oracle a taste of both extremes: euphoria when the AI contracts appeared, and anxiety when investors dug into who signed them and what they would cost to serve. This week’s numbers and commentary will sit somewhere in between those two moods.

For investors, the edge is less in guessing whether earnings per share land at 1.63 USD or 1.65 USD, and more in watching whether each quarter nudges Oracle further along the path from hopeful AI landlord to disciplined, cash generative cloud infrastructure provider.





This material is marketing content and should not be regarded as investment advice. Trading financial instruments carries risks and historic performance is not a guarantee of future results.

The instrument(s) referenced in this content may be issued by a partner, from whom Saxo receives promotional fees, payment or retrocessions. While Saxo may receive compensation from these partnerships, all content is created with the aim of providing clients with valuable information and options.