🥛 Intel’s worst day since 1974 💥

GM. This is Milk Road Stocks, the newsletter with better returns than the hours you spent doomscrolling finance Twitter.

INTEL JUST BECAME THE MEMORY BOOM'S FIRST BIG CASUALTY 💥

Intel had a monster 2026, with the stock nearly quadrupling between January and its June peak.

Then on Friday, it lost ~29% in a single session, its worst day since 1974 (yikes!).

What's nuts is the quarter itself was fine. Revenue came in at $16.5B, up 21% and matching what Wall Street expected.

The forecast is what sank it, with Intel guiding Q4 to:

$15-16B of revenue, vs. ~$17.1B expected.
$0.25 of earnings per share, vs. ~$0.43 expected.
A gross margin (the share of each sale left after paying to make the chips) of 38%, down from 44.1% last quarter.

That wiped roughly $140B off Intel's value in a day, and left the stock ~53% below its June high. 👇

Source: Nasdaq

The culprit is memory, of all things...

AI data centers are buying up every memory chip they can find, which has made memory pricier and harder to get for everyone else.

(More AI demand → pricier memory → pricier laptops and desktops → fewer people upgrading → fewer Intel chips sold.)

Intel now expects the PC market to shrink 12% this year and another 10% in 2027. Management also doesn't expect the memory crunch to ease next year.

And it stings, because Intel's data center and AI business grew 78% last quarter to $7.3B. Customers want more of its server chips, but Intel is short on factory capacity and packaging materials, and doesn't expect much relief until the back half of 2027.

Melvin (one of our PRO analysts) calls that an execution problem, and it's why he sold most of his Intel shares right after the earnings call.

He'd backed the turnaround under Lip-Bu Tan (Intel's CEO), and he's keeping a small stake in case it works. The way he put it:

"Wanting a company to succeed and believing its stock is a good investment at the current valuation are two completely different things."

(Nvidia's feeling it too, btw. Intel is its biggest stock holding, and Kyle worked out that stake is now ~$14B below its peak.)

But one company's memory problem is another company's payday...

Samsung's Q3 operating profit (profit from the core business, before interest and taxes) hit 107.4T won (~$80B), up from 12.2T won a year earlier.

That's more than double what it made in all of 2025.

Source: Samsung

Melvin pointed out that most of that came from charging more rather than selling more, with DRAM (the everyday memory inside computers and servers) prices up ~20% in a single quarter.

Samsung's entire 2027 supply of HBM (the high-end memory stacked next to AI chips) is already sold out too, and management expects shortages to last through 2028.

All four of our PRO analysts own Micron (the biggest U.S. memory maker), and Kyle and Vincent both bought more during Friday's selloff.

Kyle's note on the trade: "demand + supply shock continues."

Still, memory has always been a boom-and-bust business. Micron has dropped 50% or more ten separate times since 1995.

Shortages end when new factories come online or when buyers refuse to pay up, and Intel's guidance suggests PC makers have already hit their limit on what they'll pay.

Intel's management also expects revenue and margins to climb in 2027, and the stock is still up ~78% this year.

P.S. Melvin bought Micron at an average of ~$466 and is now up ~93%, thanks to the same shortage that just took ~29% off Intel in a day.

If you want to see what he buys next, try Milk Road PRO for a buck for 7 days.