usdaed

Enter Broadcom. They’re not building GPUs, they’re building ASICs – Application-Specific Integrated Circuits. Think of it like this: Nvidia makes a Swiss Army knife. Broadcom makes a really, really good bottle opener. It does one thing, it does it well, and it doesn’t waste space with a toothpick you’ll never use. These ASICs are faster, more efficient, and, crucially, smaller. Space in a data center is expensive, people! It’s not like they’re storing Beanie Babies in there.

You see, these chaps specialize in something called ‘semiconductors’ – tiny, almost invisible things that are the brains of everything from your telephone to the enormous, humming data centers that are popping up like mushrooms after a rain. And these semiconductors have been on a bit of a tear lately, mostly because of this newfangled craze called ‘Artificial Intelligence’ – or AI, as the cool kids call it. It’s all rather complicated, involving mountains of data and machines trying to mimic the human brain, but the gist is this: it needs lots of these semiconductors.

Let us, therefore, examine the two leading players in this grand spectacle, these purveyors of digital dreams: Taiwan Semiconductor Manufacturing and Palantir Technologies. Not because they are necessarily the most deserving of our attention, but because, alas, they are the ones most aggressively seeking it.

We’ve heard the rosy predictions, the bullish scenarios. Let’s consider the alternative. Not a crash, not a spectacular implosion. Just… a quiet settling. A slow erosion of margins, overseas expansion that sputters, and all those exciting “optionalities” failing to translate into actual, usable cash. It’s the financial equivalent of a beige room. Perfectly acceptable, but utterly forgettable.

The market, of course, is rarely concerned with the slow, quiet work of compounding interest. It prefers the spectacle. A new initiative, a daring acquisition – these are the things that capture the imagination, and briefly, inflate the price. But value, genuine and lasting, is rarely born of such fireworks. It is more often the result of consistent, unglamorous effort, a willingness to simply… wait.

To wait for the formal offering is unnecessary. The publicly available records detail the fund’s current positions, and a careful reading reveals a pattern of investment in established, if not entirely unblemished, companies. Approximately 48% of the managed stock portfolio is currently allocated to just three entities – a concentration that demands scrutiny, but does not necessarily imply recklessness.

First up: Cameco (CCJ 6.40%). These aren’t your grandfather’s uranium miners. They control 15% of the GLOBAL supply. FIFTEEN PERCENT. Kazatomprom is breathing down their neck, sure, but Cameco has the McArthur River mine… a vein of high-grade uranium so rich, it practically bleeds power. They hauled in $3.4 billion last year, a solid 11% jump. And let me tell you, in this insane market, uranium was the ONLY energy resource that wasn’t circling the drain. A 16.9% profit margin? In mining? That’s witchcraft, pure and simple. But they don’t just dig stuff up. They refine it, fuel it, and they own 49% of Westinghouse, the guys building the AP1000 reactors. These aren’t your daddy’s reactors, either. We’re talking advanced, efficient… practically foolproof. (Don’t quote me on that last part.) Cameco has a hand in EVERY stage of this madness. From the mine to the meltdown… I mean, the power plant.

But the cough is significant, because the numbers are… well, they’re showing off. The Shiller P/E ratio, or CAPE, as the professionals call it—a name that sounds suspiciously like a type of outerwear—is hovering around 39.2. That’s nearing 2000 levels, which, if you recall, was a time when people were convinced pets.com was a viable business model. It’s a ratio that looks at ten years of earnings, smoothing out the bumps, and right now, it’s screaming “expensive.” My father used to say anything over 20 was a fool’s game, but he also wore socks with sandals, so…

The price of oil, you see, has been doing a jiggly dance, shooting upwards like a startled jack-in-the-box. Naturally, the people at UPS started to fret about fuel costs. But here’s a secret: fuel isn’t quite the monster they make it out to be. In 2025, all the fuel used to whizz parcels about cost a measly $4.3 billion. That’s a lot of money, of course, but only 5.3% of their total expenses of $80.8 billion. A mere crumb, really.

He’s been shedding weight – Meta, Amazon, even Microsoft taking a hit. Not a fire sale, mind you, more like a preemptive strike. A seasoned gambler knows when to cut his losses, or at least redistribute the risk before the whole thing goes nuclear. But the real story isn’t what he dumped, it’s where he doubled down. NVIDIA. That’s where the man’s conviction lies. A cold, hard, silicon-fueled conviction. The kind that keeps you up at night, staring into the abyss of exponential growth. And it’s terrifying, frankly.