I think Oracle is one of the most interesting speculative situations in tech right now, but speculative is the key word. The stock has lost more than 60% from its all-time high of $345 in September 2025. It's been stabilizing since early February. The earnings report on March 10th gave it a brief lift. Since then, it's drifted back down. My read: the short-term uptrend is intact, which makes it an interesting trade. As a long-term investment, Oracle hasn't convinced me yet.

Some context: the nearly 200% rally between April and September 2025 was driven by AI and cloud enthusiasm – specifically OCI, Oracle's cloud infrastructure for data-intensive AI workloads. The cloud business is genuinely growing fast: up 35% in the first nine months of the current fiscal year, to $24 billion. The backlog (remaining performance obligations) has exploded by 325% to $553 billion. Impressive numbers. Hard to argue with. Or is it?

To meet demand, Oracle has to spend heavily. Capital expenditures jumped from $6.9 billion in fiscal year 2023/24 to $21.2 billion in 2024/25. The forecast for 2025/26 is around $50 billion. The problem: that kind of spending can't be funded from cash flow alone. So debt is piling up. Net debt climbed from $78.5 billion to $123 billion in twelve months. Credit default swap spreads hit their highest level since 2009. Oracle became an involuntary proxy for AI credit risk.

In February, Oracle laid out its financing plan, which resolved some of the uncertainty. CDS spreads eased, the stock stabilized. I see this as a technical recovery, not a trend reversal.

First: a sustained break above the March 11th interim high of around $172. That would confirm the short-term uptrend. Second: clarity on Capex going forward. By June, when Oracle reports its full-year 2025/26 results, I need to see where the investment cycle is heading. If another wave of debt is coming, the leverage concerns will be back.

The third risk, and one that doesn't get enough attention, is customer concentration. More than 54% of Oracle's backlog comes from a single customer: OpenAI. That's a dangerous dependency. A potential OpenAI IPO could actually be a positive catalyst for Oracle if it brings in fresh capital. But what happens if OpenAI runs into financial trouble?

There's also structural margin pressure: the gross margin is declining because cloud infrastructure is less profitable than Oracle's traditional software business. That trend isn't over.

And there's a related risk that's easy to overlook: depreciation. The billions Oracle is pouring into infrastructure today will show up as depreciation charges tomorrow. Tech hardware ages fast. As those assets are written down, earnings will come under additional pressure, quietly but persistently.

Oracle isn't a bad company. It's a good company in an expensive transformation with an uncertain outcome. Over the next few weeks, with sentiment staying positive, the stock can run. Over the next few years, I need more clarity on Capex, margins, and the OpenAI dependency. That clarity probably won't come before summer.

Disclaimer: This newsletter is for informational purposes only and does not constitute investment advice. I am not a financial advisor. Always do your own research before making any investment decision.

Disclosure: I may hold direct or indirect positions (including options) in any securities mentioned in this newsletter. My opinions are my own and always honest.