It’s like ordering room service every night for a month. It’s great, until you get the bill. And then you’re thinking, “Maybe I should have just gone to the grocery store.” That’s where we are with these companies. And the worst part is, it’s 2026, and they’re still expecting a miracle? It’s just… exhausting.
The cloud computing division, Amazon Web Services (AWS), is the engine driving this renewed optimism. It is a vast, complex undertaking, offering businesses the tools they require to function in the digital age. More recently, it has become a focal point for those developing AI applications, drawn by the scale of its data centers and the associated infrastructure. This is not progress for its own sake; it is simply a matter of providing a service for which there is demonstrable demand.
They make the da Vinci surgical robot. A remarkably precise device, if you ignore the fact that it’s named after a man who spent most of his time sketching flying machines and worrying about anatomy. They’ve been at it a while, ending 2025 with over 11,106 of these contraptions installed globally. That’s a 12% increase, which, in the grand scheme of things, is roughly equivalent to adding another small kingdom to the robot’s domain. More importantly, the number of actual surgeries performed by these robots rose 18% year over year. People, it seems, are increasingly willing to let a machine poke around inside them.1
That said, not all ships are created equal. Some are sturdy, built to weather any storm, while others… well, let’s just say they’re held together with optimistic varnish. I have a few holdings that would allow me to sleep soundly, even if the market decided to imitate a startled flock of pigeons. It’s not about avoiding a dip, it’s about enduring it with a degree of composure.
Three billion souls gazing into the Facebook and Instagram mirrors – a truly staggering number. It’s a captive audience, and Meta Platforms, clever fellows that they are, are milking it for all it’s worth. They’ve discovered that a well-placed advertisement, guided by the all-seeing eye of AI, is far more effective than shouting into the void. It’s simple, really: show people what they want before they know they want it. The returns, as you might imagine, are quite substantial. A 26% year-over-year revenue increase? Why, that’s enough to make a bureaucrat blush.
Okay, let’s be clear. The valuation isn’t the problem problem. It’s the starting point. And right now, it’s a high one. The S&P 500 is doing alright for itself, hovering near all-time highs, but its P/E is a comparatively sensible 28. The iShares U.S. Pharmaceuticals ETF? A modest 30. Lilly’s at 53. Do the math. I’m not a mathematician, but even I can see that’s a significant gap. It’s like showing up to a potluck with a diamond-encrusted soufflé. Impressive, sure, but a bit… much.
Two years ago, with Intel trailing TSMC in manufacturing prowess by a margin that felt… significant, such a feat would have been considered, shall we say, statistically improbable. (Think of it like trying to predict the exact arrangement of tea leaves after a particularly vigorous stirring. Possible, yes, but not something you’d base a quarterly earnings forecast on.) But thanks, in part, to Intel’s 18A manufacturing process, Panther Lake appears to be a much-needed inflection point. A home run, if you will, though one executed with a rather complex series of physics and material science.
Amongst my own holdings, two companies presently claim my particular esteem: Costco Wholesale and Roku. These are not merely matters of financial interest, but of observing a certain… competence in their respective spheres. They possess, in my estimation, a capacity for exceeding expectations, and thus warrant a closer scrutiny.
They produced eighteen thousand, three hundred and seventy-eight vehicles this past year. A hundred and four percent increase, they announce. A significant number, to be sure. One can almost hear the hopeful cadence in their reports. But numbers, like people, require context. A percentage increase is easily achieved from a small base. It’s like a child growing an inch – a triumph for the child, perhaps, but hardly a challenge to a grown man. The capital required to birth these machines is immense, a burden that weighs heavily on any nascent enterprise. The ramp-up, as they call it, is a slow, arduous climb.
The company itself managed to exceed its own modest expectations, a feat rarely celebrated in polite society. Revenue and profits, while not exactly soaring, were at least respectable. Yet, the underlying trend remains stubbornly downward. Subscribers, like moths to a dying flame, continue to drift away. A most inconvenient truth.