Oracle's AI Snapshot: The Tech Trade Nobody is Watching

Okay, so this one actually surprised me.

If you've been watching the financial news feeds today, you're probably seeing a whole lot of panic about macro events, futures taking a dive, and the usual doom-and-gloom commentary. But while everyone is busy looking out the front window at the macro storm, fourth-quarter earnings have quietly slowed to a trickle. And sitting right there, almost completely ignored by the mainstream retail crowd, is Oracle.

Yes, Oracle. The database company that your company's IT department has been complaining about since 1998.

Oracle is stepping up to the plate as the main earnings event this week, and they are about to offer us the clearest snapshot of the AI trade we've seen all year. Which is wild.

Think about it. We spent the last two years obsessing over chipmakers. If a company even whispered the word "semiconductor," their valuation went to the moon. But we are past the initial hardware hoarding phase. Now, we are in the deployment phase. Companies have the chips – now they have to actually build the data centers, string the infrastructure together, and run the models.

And I'll be honest – this one surprised me. Oracle has somehow maneuvered itself into being one of the most critical bellwethers for whether this AI infrastructure build-out is actually generating cash or if it's just a massive capital expenditure black hole.

Let's talk about what this means practically.

When you look at the enterprise software space, you have the big three cloud providers: Amazon Web Services (AWS), Microsoft Azure, and Google Cloud. For a long time, Oracle was viewed as the legacy dinosaur trying to play catch-up. But in the AI era, their Gen2 Cloud infrastructure is specifically designed for high-performance computing. They partnered heavily with Nvidia early on to build superclusters.

If you want to know if corporations are actually spending money to train and deploy AI models right now, you don't ask the people making the apps. You ask the people renting out the heavy machinery.

Look at those numbers. The legacy on-premise stuff is practically flatlining, but the cloud infrastructure growth is where the real story is hiding.

Now here's where it gets interesting.

Oracle's earnings aren't just about Oracle. They are a proxy for the entire secondary tier of the tech market. If you read my recent piece on Why Oracle's Sell-Off Could Be The Buy Of The Year, you know that Wall Street has a habit of mispricing companies that are undergoing massive internal transitions.

We love a clean narrative. We want a company to either be a hyper-growth startup or a stable dividend payer. But what happens when a massive, mature company pivots hard into hyper-growth territory in a specific division? The market gets confused.

Investors are terrible at pricing in slow, structural shifts. We want the instant gratification of a flashy product launch. But the real money in tech isn't usually made on the flashy app; it's made on the boring, invisible plumbing that makes the app work.

Have you ever tried migrating a massive corporate database from an on-premise server to the cloud?

I haven't personally, because I value my sanity, but I have friends in enterprise IT who tell me it's a nightmare of epic proportions. Once a company moves its data into a cloud ecosystem like Oracle's or Azure's, they do not leave. The switching costs are astronomically high.

So when Oracle reports their cloud infrastructure revenue this week, we aren't just looking at a three-month snapshot. We are looking at sticky, recurring revenue that will likely stay on the books for the next decade.

But wait – there's more to this.

While we're talking about the infrastructure of the internet and how AI is changing the plumbing, we have to talk about what's happening on the consumer-facing side of the web.

Today, The Trade Desk's CEO came out with a fascinating perspective, essentially saying that the open internet just got stronger.

For years, the digital advertising narrative has been dominated by the "walled gardens" – primarily Meta and Google. The conventional wisdom was that if you wanted to reach consumers effectively, you had to pay the toll to the tech giants because they had all the user data and the best algorithms. The open internet – which is basically every news site, blog, streaming service, and independent publisher out there – was viewed as the messy, inefficient alternative.

But AI is completely flipping that script.

Here's what I actually think about this...

The Trade Desk operates as a demand-side platform (DSP). They help ad agencies and brands buy ads across the open internet programmatically. In the past, targeting an ad on a random cooking blog was hard because you didn't have the rich user profile that Facebook had.

But with modern AI, you don't necessarily need a creepy, hyper-specific dossier on every single user. AI can analyze contextual signals – the time of day, the exact content of the article, the pacing of the user's reading, the device they are using – and predict with incredible accuracy whether that person will click an ad for a new SUV.

AI is acting as the great equalizer. It is taking the messy, unstructured data of the open internet and turning it into a highly efficient marketplace that rivals the walled gardens.

Going a step further...

When you combine Oracle's infrastructure play with The Trade Desk's open internet play, you start to see the real shape of the 2026 tech economy.

We are moving away from the centralized monopolies. Yes, the mega-caps are still dominant, but the underlying technology of AI is allowing secondary players to punch way above their weight class. If you have the right infrastructure (like Oracle is providing) and the right AI-driven models (like The Trade Desk is utilizing), you can build incredibly profitable businesses without having to own the entire ecosystem.

This is the part that genuinely worries me about how most retail investors are approaching the market right now.

People are still trying to play the 2023 playbook. They are looking for the next Nvidia. They are looking for the one stock that is going to revolutionize everything overnight.

But that's not how mature technological shifts work. Think back to the transition from desktop to mobile. Apple made a fortune on the iPhone, sure. But the real wealth was generated over the next decade by companies that figured out how to use mobile connectivity to build entirely new business models – Uber, Airbnb, Instagram.

We are in the "Uber/Airbnb" phase of the AI trade. We are past the invention of the smartphone, and we are now figuring out what to build on top of it.

If you want to understand how this is quietly reshaping index funds, you should definitely check out my deep dive on The Hidden S&P 500 Shakeup: Why AI Just Hijacked Your Index Fund. The companies that are actually benefiting from this shift aren't always the ones making the headlines.

Okay so real talk for a second.

I spent an embarrassing amount of time this weekend reading through enterprise software earnings call transcripts. I know, I need a life. But the recurring theme across all of them was efficiency.

CEOs are no longer talking about AI as a fun science experiment. They aren't issuing press releases about cool chatbots. They are talking about AI as a margin-expansion tool. They are using AI to write code faster, to optimize their supply chains, to buy advertising more efficiently, and to reduce their server costs.

This is why Oracle's earnings this week are so critical. If Oracle shows a massive beat on their cloud infrastructure revenue, it tells us that corporate America is still heavily investing in the foundational tools required to achieve that efficiency.

If they miss? Well, that's a different story. A miss would suggest that companies are hitting the pause button, digesting the massive investments they've already made, and waiting to see some actual return on investment before they sign another billion-dollar cloud contract.

And this is exactly why the market is so incredibly tense right now.

We are at a transition point. The narrative is shifting from "growth at all costs" to "show me the money." Investors want to see the ROI on AI.

When The Trade Desk CEO says the open internet is getting stronger, he is essentially saying, "Our AI investments are showing up in the bottom line of our clients. We are generating better returns on ad spend than we were a year ago."

When Oracle reports, we need to hear a similar message. We need to hear that their clients aren't just buying cloud compute space because of FOMO, but because it is actively improving their business operations.

Here's the part that actually matters.

You don't have to be a tech analyst to make money in this environment, but you do have to be willing to look past the front page of the financial news.

Right now, everyone is fixated on the macro terrors of the day. And don't get me wrong, those things matter. But the structural changes happening in the economy—the way businesses operate, the way advertising is bought and sold, the way data is stored and processed—those changes don't stop just because the futures market is having a bad morning.

The companies that provide the picks and shovels for the digital economy are quietly building massive, durable moats. Oracle has spent years being the butt of jokes about legacy software, but they are currently sitting on some of the most highly sought-after AI server capacity in the world. The Trade Desk has spent years fighting against the absolute dominance of Google and Meta, and they are finally seeing AI level the playing field in their favor.

My honest take: The most profitable investments over the next three years aren't going to be the companies that invent a new AI model. They are going to be the boring, profitable companies that figure out how to use existing AI models to drastically lower their customer acquisition costs and improve their operational efficiency.

So, when Oracle drops their numbers this week, don't just look at the top-line revenue beat or miss. Look at the cloud infrastructure growth rate. Look at the capital expenditure guidance. Look at the comments about customer demand for AI superclusters.

That's where the real signal is. Everything else is just noise.