Two days ago the semis fell 11% and the financial press reached for the bubble headlines. Then Thursday showed up and quietly mugged the narrative: the PHLX semiconductor index rose 3.06%, back to up 83% on the year — and the spark wasn’t the usual star. It was Micron (MU), which jumped 4.5% after unveiling a $250 billion U.S. manufacturing plan, a number that used to be a decade of capex for the entire industry. The rest of the memory bench followed it out the door: Sandisk (SNDK), Western Digital (WDC), ON Semiconductor (ON), Marvell Technology (MRVL), Advanced Micro Devices (AMD). What sat out is the real tell — Nvidia (NVDA), Alphabet (GOOGL) and Microsoft (MSFT) barely moved. The market is re-picking its AI winners, and for now it’s paying the suppliers before the celebrities.
The logic is old and blunt. Nobody knows which model or assistant wins the decade; everybody knows the buildout eats staggering amounts of memory, and memory pricing is the tightest bottleneck in the chain. So investors are starting to ask, with real money, when the hyperscalers’ estimated $710 billion of AI spending actually turns into profit — and rotating toward the names being handed that check. SK Hynix lists in the U.S. today at roughly six times forward earnings, which is either the cheapest seat in the theater or a warning label; memory is the most boom-and-bust corner of tech, and six-times is what the market charges for that history. Here’s the part a market column won’t tell you, and it’s the whole planning point: the real decision isn’t “do I believe in AI,” it’s how much of your retirement rides on it. That’s the Investments & Risk question (IN04) — concentration and position sizing. As a general planning principle, an AI sleeve should be built so an 11% chip week is a headline, not a lifestyle change. If you own the theme through the giants alone, or you own nothing and feel the itch to fix that in one afternoon at week-one-of-a-rotation prices, it’s the sizing that deserves the second look, not the ticker.
You don’t wait for the storm to learn the roof leaks — you glance at the forecast and go check it on a dry afternoon. Same with a portfolio: a review is fifteen minutes, bring your latest statement, and we’ll see whether your AI exposure is sized for the week the chip index drops another 11% and you shrug. If the honest answer is “nothing much happens to my plan,” you’re already done. If you’re not sure, that’s the fifteen minutes.