Just caught something interesting from Nvidia's latest earnings call that probably deserves more attention than it's getting. Jensen Huang basically dropped some wild numbers about where AI infrastructure spending is headed, and it completely reframes how you should be thinking about the company right now.

So here's the thing - Nvidia is shipping the Vera Rubin platform starting in the second half of this year, and the performance jump is genuinely insane. We're talking about training AI models with 75% fewer GPUs compared to their current Blackwell chips, plus slashing inference token costs by 90%. That's not incremental improvement, that's a different category of efficiency.

But the real story came during Jensen Huang's comments to investors. He pointed out that the world historically spent around $400 billion annually on classical computing infrastructure. Then he casually mentioned that AI workloads require about a thousand times more computing capacity. Let that sink in for a second. A thousand times.

Huang previously estimated AI data center spending could hit $4 trillion per year by 2030. At the time that sounded ambitious, but now it's starting to feel less like speculation and more like baseline math. Especially since bringing down inference costs is going to unlock way more usage across the board.

On the valuation side, here's what caught my eye: Nvidia's trading at a P/E of 36.1 right now, which is actually 41% below its 10-year average of 61.6. That's a significant discount to where it normally trades. Wall Street's consensus for fiscal 2027 earnings is $8.23 per share, putting the forward P/E at just 21.5. For context, the S&P 500 is sitting at a trailing P/E of 24.7, so Nvidia would actually be cheaper than the broader market.

Given the scale of what Jensen Huang is describing - this massive expansion in AI infrastructure needs - I'm skeptical the stock stays cheap for long. If those earnings estimates hit, the stock would need to more than double just to get back to its historical P/E average. Even a fraction of that upside would be a solid move.

Worth keeping an eye on, especially if you're thinking about the longer-term AI infrastructure play.