OpenAI’s Bets: Microsoft or Oracle?

OpenAI, you see, is spending money like a drunken sailor on shore leave. A lot of money. Billions. Enough to make Croesus blush, and then probably weep. They’re building a future powered by thinking machines, and that future requires… well, a lot of electricity. So it goes.

They’ve made deals with two giants: Microsoft and Oracle. Both are building the server farms, the digital real estate, for OpenAI’s ambitions. It’s a bit like building castles in the air, only these castles need cooling systems and a whole lot of GPUs. A truly modern absurdity.

Microsoft, bless their corporate heart, has been in this game longer. They saw something in OpenAI back in 2019, before the chatbots started writing poetry and terrifying marketers. Now they own a piece of it – 27%, if you’re keeping score – and have a commitment from OpenAI worth a quarter of a billion dollars. A tidy sum, even for them.

Oracle, late to the party but eager to dance, has a deal worth even more – three hundred billion dollars. That’s a number that makes your head spin, isn’t it? It’s like trying to count all the grains of sand on a beach. Pointless, really. But they’re building data centers, too, hoping to catch the OpenAI wave. A gamble, if you ask me. And I often do.

How Does One Pay for All This?

OpenAI is promising to spend over half a trillion dollars with these two alone. Plus another eight hundred billion with various chipmakers and cloud providers. They’re currently making about twenty billion a year. Do the math. It doesn’t quite add up, does it? They’re relying on vendor financing, stock warrants, and a healthy dose of optimism. It’s a precarious arrangement. Like building a house of cards during an earthquake.

Some folks complain about these circular deals. They say it’s unsustainable. Maybe they’re right. But if OpenAI succeeds, everyone wins. If not… well, the whole thing could come crashing down. And that would be a sight to see. A truly magnificent disaster.

Who’s Paying the Builders?

Microsoft and Oracle are both throwing money at infrastructure. New data centers, mountains of GPUs, cooling systems that could freeze a small planet. Microsoft spent almost thirty-five billion dollars last quarter. Oracle tripled its spending. It’s a race to build the future, and it’s expensive. Very expensive. So it goes.

Microsoft, being the behemoth it is, can afford it. They have cash flowing out of their ears. Oracle, on the other hand, is borrowing money to pay for it. They’re betting that OpenAI will pay them back. A risky bet, wouldn’t you say? They’re also burning through cash at a frightening rate. Ten billion dollars last quarter. It’s like setting money on fire, only with more servers.

Microsoft’s profit margins might take a hit, but they have a diversified business to fall back on. Software, cloud services, video games… they’re not entirely dependent on OpenAI. Oracle, however, is putting all its eggs in one basket. A dangerous game. If OpenAI stumbles, Oracle could be left holding the bag. A very heavy bag.

Oracle trades at a premium, a price-to-earnings ratio of 28. Seems a bit rich for a company so heavily reliant on one customer. Microsoft, at 29.5, is only slightly more expensive. And they have a more diversified business, a stronger balance sheet, and a history of success. It’s a simple choice, really. A bit like choosing between a sturdy oak and a fragile sapling. So it goes.

Investing, of course, is always a gamble. But some gambles are more sensible than others. And in this case, the sensible choice seems clear. Microsoft is the safer bet. The less likely to leave you staring into the digital abyss. And isn’t that what we all want, really? A little bit of peace of mind in a chaotic world.

2026-01-18 17:52

How Does One Pay for All This?

Who’s Paying the Builders?

Read More