Nvidia: A Most Peculiar Ascent

by

The air hums with it, you see. Not electricity, though there’s plenty of that. It’s the hum of expectation. For the last few years, the whispers have grown into a roar, all centered on Artificial Intelligence. A clever bit of thaumaturgy, really – imbuing clockwork and crystal with the semblance of thought. It promises riches beyond imagining, and the potential to automate away the truly tedious bits of existence. Though, naturally, someone will find a way to make even that tedious.

And at the heart of this… well, let’s call it the Cognitive Revolution… sits Nvidia.1 A company that, not so long ago, was known for making things that made pictures appear on glowing rectangles. Now? They’re practically the alchemists of the age, transmuting silicon into… well, more or less everything. Their market value has swelled to a truly alarming degree – more than four trillion crowns since the beginning of 2023. A figure that, frankly, makes one suspect someone has misplaced a decimal point.

Nvidia’s ‘graphics processing units’ – or GPUs, as the initiates call them – are the preferred engine for these thinking machines. Their current chips – Hopper, Blackwell, and the frankly intimidating Blackwell Ultra – are, by all accounts, rather good at what they do.2 And Jensen Huang,3 the company’s… let’s call him ‘Chief Artificer,’ seems determined to churn out a new marvel every year. A pace that’s either genius or sheer recklessness, depending on your point of view.

But history, as any seasoned librarian will tell you, is littered with the wreckage of ‘sure things.’ And even the most impressive ascents are often followed by… adjustments. So, let us consider, with a slightly cynical eye, the five potential snags that might yet trip up Nvidia’s parabolic climb in the coming months.

1. The Inevitable Puff of Smoke

The first, and perhaps most obvious, danger is the bursting of the ‘AI bubble.’ Every new grand scheme – the internet, genome decoding, even those ridiculous self-folding laundry machines – goes through this phase. A period of breathless hype, followed by a rather undignified collapse. Investors, you see, are remarkably bad at judging the difference between a genuine innovation and a particularly shiny distraction.

The problem is, truly revolutionary technologies take time to mature. The internet, for instance, didn’t suddenly transform society overnight. It took years of clumsy interfaces, dial-up modems, and questionable websites before it became the ubiquitous force it is today. And even now, some argue it’s mostly used for cat videos.

Currently, businesses are throwing money at AI with the enthusiasm of a dwarf at a gem auction. But they haven’t yet figured out how to actually profit from it. The return on investment, shall we say, is… elusive. Which suggests that we may be heading for a rather spectacular deflation of expectations.

And if that happens, Nvidia, as the current poster child for AI, is likely to feel the pinch most acutely.

2. The Internal Guild Wars

Competition is, of course, the natural order of things. But sometimes, the most dangerous rivals aren’t external. Several of the ‘Magnificent Seven’ – those powerful corporations who seem to control everything – are developing their own AI chips.4

Nvidia has done a clever job of locking in its customers with its CUDA software platform.5 A proprietary system that makes its GPUs particularly attractive. But these corporate giants aren’t known for their loyalty. They’d much prefer to build their own solutions, even if they’re slightly less efficient, than remain dependent on a single supplier.

If these internal efforts gain traction, it could reduce the scarcity – and the pricing power – that Nvidia currently enjoys.

3. The Dragon’s Shadow

Nvidia has a… complicated relationship with China. For years, the Middle Kingdom has been a crucial market. But geopolitical tensions have thrown a spanner in the works. The U.S. government has restricted exports of high-powered chips, and tariffs have added to the complications.

There was a brief glimmer of hope when the Trump administration allowed Nvidia to ship its H200 GPUs to China, in exchange for a 25% tax. A rather steep price, to be sure, but a price worth paying, apparently. Except… Chinese regulators aren’t allowing the chips into the country.6

The longer this stalemate continues, the more likely it becomes that China will develop its own alternatives to Nvidia’s technology.

4. The Artificer’s Dilemma

Jensen Huang is, undeniably, a visionary. His relentless pursuit of innovation has propelled Nvidia to the forefront of the AI revolution. But there’s a risk that his ambition is… unsustainable.

Introducing a new GPU every year risks rendering older models obsolete. This could delay upgrade cycles, as businesses wait for the next, even more powerful chip. Or, worse, it could entice them to trade down to cheaper, older models.

In other words, Huang’s desire to maintain Nvidia’s dominance could backfire.

5. The Market’s Fickle Fancy

Finally, there’s the simple fact that the stock market is… well, a bit mad. Nvidia’s price-to-sales ratio is currently hovering around 30.7 A level that hasn’t been seen since the dot-com bubble. And the S&P 500’s Shiller P/E ratio is even more alarming.8

History suggests that companies with such high valuations are vulnerable to a correction. If the market as a whole were to decline, Nvidia would likely be hit particularly hard.

So, there you have it. Five potential snags that could derail Nvidia’s parabolic climb. Whether any of them will actually materialize remains to be seen. But as any seasoned observer of the market knows, fortune is a fickle mistress. And even the most impressive ascents are rarely without their bumps.

1

Nvidia, of course, isn’t

really

an alchemist. Though some of their marketing materials might suggest otherwise.

2

The exact specifications of these chips are, naturally, shrouded in secrecy. Which is a good thing, really. You wouldn’t want just anyone building a thinking machine.

3

Huang is, by all accounts, a rather eccentric individual. He’s often seen wearing a leather jacket, which is unusual for a CEO.

4

The ‘Magnificent Seven’ are Apple, Microsoft, Alphabet (Google), Amazon, Nvidia, Tesla, and Meta (Facebook). They control a disproportionate share of the market, which is a bit worrying, to be honest.

5

CUDA is a proprietary software platform that allows developers to write code for Nvidia’s GPUs. It’s a clever bit of engineering, but it also creates a lock-in effect.

6

The reasons for this delay are complex and involve a great deal of political maneuvering. It’s enough to say that things are… complicated.

7

The price-to-sales ratio measures a company’s market value relative to its revenue. A high ratio suggests that the stock is overvalued.

8

The Shiller P/E ratio is a measure of the market’s overall valuation. It’s calculated using average earnings over the past ten years, which smooths out short-term fluctuations.

Read More

2026-01-20 12:13