Stocks hit an all-time high this week. Nobody noticed.

Dear Fellow Investor,

Stocks hit an all-time high today.

Miss it?

So did Wall Street.

Because while the S&P 500 is down this month…

An index tracking the same S&P 500 stocks on an equal-weighted basis just hit its highest level, ever.

While chip stocks, the most crowded trade in history, were falling apart, the average American stock was hitting a record high.

When Nvidia sneezes, the regular S&P 500 index catches a cold.

The equal-weighted version treats every company the same.

Coca-Cola counts just as much as Nvidia does.

And today, that version closed at an all-time high…

At the same time a basket of the biggest semiconductor names dropped 6%.

Nine of the eleven S&P sectors finished higher today. Consumer staples rose 3.7%. Healthcare added 3%. Materials climbed 2.2%.

And Apple pushed straight through a $5 trillion market cap ahead of its earnings.

Now here's the other side of the ledger.

The semiconductor index is down more than 23% this month.

The Nasdaq 100 is in a 10% correction.

So which is it?

You may feel like the market is falling apart…

And yet, it’s setting records.

Both things are true at the same time.

Here’s the truth:

The money isn’t leaving. It's moving.

What Is Actually Being Sold

This is where most people get it wrong. They see red chip screens and think, "the market has decided AI is over."

The market has decided nothing of the kind.

Look at what's actually being dumped, and you'll see three very specific things — not one big idea.

First, memory.

Micron. Sandisk. Samsung. SK Hynix.

These companies sell a commodity. When memory prices rise, they print money. When prices fall, they don't. That has been true for forty years and it will be true for forty more.

That's not a referendum on artificial intelligence. That's a cycle doing

what cycles do.

Second, the equipment makers.

The machines that etch these chips were a near-monopoly. Now there are reports out of China of homegrown machines that compete.

Again — not a verdict on AI. A verdict on who gets to sell the tools.

Third, and this is the big one. The vendor itself.

Over the weekend, the Wall Street Journal reported that Nvidia is in talks to guarantee roughly $250 billion so that OpenAI can lease a giant new data center campus in Ohio.

Why would Nvidia need to do that?

Because OpenAI can't borrow that kind of money on its own credit. So Nvidia's balance sheet would stand in for OpenAI's.

There are separate talks reported for up to $350 billion more to help finance the chip purchases themselves.

Now, nothing is signed. These talks are early. They could fall apart or change completely, and I want to be fair about that.

But the market got the message anyway.

The company selling the shovels is now being asked to co-sign the loan for the fellow doing the digging.

Folks, that is not an AI problem.

That is a position problem.

And there's a difference.

The Cisco Story Nobody Finishes

You are going to hear a lot of smart people this week tell you to sit

tight. And most of them will bring up Cisco.

They're half right.

Let me give you the half they always leave out.

March 27, 2000. Cisco Systems closes at $80.06 a share. That same day, it passes Microsoft to become the most valuable public company on the face of the earth.

And here's the thing — Cisco was not a fraud.

Cisco was right.

The internet was real. The traffic was real. It ran through their

hardware, exactly like the bulls said it would.

Then the stock fell about 90%.

It bottomed out in late 2002.

And it did not climb back to that $80 until December 10, 2025.

Twenty-five years. Eight months. Thirteen days.

Over that same stretch, Cisco's revenue nearly quintupled. Profits

quadrupled. The business delivered everything it promised.

The people who bought at the top still lost money to inflation for a generation.

That's the lesson, and it has nothing to do with patience.

Being right about the technology is not the same thing as being right about what you own.

Tomorrow

Two things land Wednesday.

At 2 p.m. Eastern, the Federal Reserve tells us what it's doing.

The funds rate sits at 3.50% to 3.75% right now. Futures traders put the odds of a hike at roughly 36% — up from about 16% just a week ago.

A hike is not the base case. But under Chair Warsh it is a live question, and that's new.

The reason is sitting in your gas tank.

Oil topped $100 a barrel last week before easing back.

And this morning we learned consumer confidence slipped again, to

90.8 from 92.2.

How folks feel about conditions right now has fallen three months in a row.

Then, after the closing bell, a company I've been watching for an entirely different reason reports its quarter.

It doesn't sell memory chips.

It doesn't sell the machines that make chips.

And it isn't guaranteeing anybody's construction debt.

It licenses the design that chips get built on. It gets paid a small cut when those chips ship — and it collects that cut no matter which company's name ends up on the package.

Volume, not price. Royalties, not credit.

That is a very different seat to be sitting in on a day like today.

We put together a full briefing on this business, how the royalty model actually works, and what tomorrow night's numbers could mean for it.

Whatever you decide to do with it, do this much: stop asking whether AI is real.

That's the wrong question, and it's been the wrong question all month.

Ask which layer of it you're standing on when the bill arrives.

“The Buck Stops Here,”