THE DEAN’S LIST: Chips Show Some Cracks

For weeks, I have been pointing at one risk above all others. The whole market was leaning on a handful of chip names. Last week, that trade finally cracked, and the chips gave back some meaningful gains. Keep in mind, they doubled in less than 90 days, from March to July. Before you decide the sky is falling, pull up the Single-Digit Millionaire portfolio and look at how it is balanced across stocks, cash, gold, and a little bitcoin. A week when the hottest trade on earth sells off, and the war pushes oil higher, is exactly the week a plan built for both directions earns its keep. The screens were red. The structure held.

Dean’s note:
The thing I have been warning you about for a month finally happened. The AI trade cracked.

The chip stocks, the crowded corner that has led this entire market, fell into a “bear market” last week. The main chip gauge dropped about 20 percent from its record high, its worst week since early 2025. Remember, this is a group that had run up more than 100 percent. A rally that stretched has finally snapped back, hard. The Nasdaq led the market lower, and the S&P closed its first losing week in three.

So what broke the spell? A headline out of China. A startup called Moonshot released a new artificial intelligence model that it claims performs on par with the best from OpenAI and Anthropic. That one claim prompted investors to ask whether the whole boom was built to avoid. If a small team can match the giants for a fraction of the money, is the trillion-dollar spending arms race really worth it? Add in a report that Alphabet's next big model is running months behind, and valuations stretched to the moon, and you have all the fuel a selloff needs.

But let me show you the part that keeps me calm. In the very same week, TSMC, the company that actually makes the world's most advanced chips, posted a record profit. Revenue up 36 percent. Earnings up more than 70 percent. The demand for these chips is not slowing down. It is accelerating. So this was not the AI story breaking. It was the AI price getting a long-overdue haircut. A great company and a great stock are not the same thing, and last week the market remembered that in a hurry.

And notice what did not happen. The money did not all run for the exits. The rest of the market held up far better than the Nasdaq, including small caps and the equal-weighted S&P 500, while the chips got smoked. This has the look of the crowded trade correcting while the rest of the market holds its ground. That is rotation, not ruin, the same movie I described back in June, just louder and meaner this time.

Here is the one thing that could turn this from a healthy correction into a real problem. Oil. While the chips were falling, the war got worse. The US reimposed its naval blockade, Iran hit two more tankers in the Strait of Hormuz, and crude surged more than 14 percent on the week, past 82 dollars. A chip correction, the market can digest. A chip correction plus an oil shock is a different animal.

So here is where I land. This is a healthy, overdue correction in the most crowded trade on earth, not the end of the bull market. But this time it is a real correction, not the breather I described a few weeks ago. Stay invested because the demand for these chips is still real, and the money is rotating, not fleeing. Stay selective, because the days of paying any price for anything with AI in the name are over. And watch the barrel, because the oil is the wildcard that decides how this ends.

A few days ago, the hottest trade on earth finally cracked. Here is how it played out.

Monday (July 13): The war was back on the front page. Oil surged and chip stocks tumbled as Iran hit ships in the Strait of Hormuz and the US struck back. The S&P slid about 0.8 percent to start the week on the back foot.

Tuesday (July 14): The big test. June inflation cooled to 3.5 percent, below expectations, and the big banks kicked off earnings with a bang, with Goldman jumping 9 percent. But IBM cratered 26 percent on a rare profit warning, its worst day on record. A split tape.

Wednesday (July 15): The chips started to roll over. Alphabet and the semiconductor names sold off, producer prices landed, and the mood soured.

Thursday (July 16): The selling spread. The Nasdaq fell more than 1 percent as the chip rout deepened. Retail sales came in soft, up just 0.2 percent. And TSMC posted a record profit that the market barely noticed.

Friday (July 17): The break. The chips fell into a bear market, down about 20 percent from their peak, after the Chinese startup unveiled its new AI model. Oil surged again, with Brent up nearly 5 percent on the day. The S&P closed its first losing week in three.

Monday (July 20): The dust is settling on the worst week for chips since early last year, with oil near its highs and the AI trade on its heels. The focus now turns to the biggest tech names, which report over the next two weeks, and whether they still plan to spend the fortune they promised on AI.

For two years, the chip stocks were the market. They led every rally, doubled and doubled again, and dragged the whole index up behind them. Last week, that ended, at least for now. The main semiconductor gauge dropped about 20 percent from its record high, the official line for a bear market, and logged its worst week since early 2025.

The selling was broad. Applied Materials, Lam Research, Intel, KLA, and Arm all fell about four percent in a single session. Even Nvidia and Micron, the generals of the whole move, dropped by more than 2%. A group that was up more than 100 percent finally found the ceiling, and the crowd headed for the door all at once.

Dean’s note:
I have watched many crowded trades over the years, and they almost always end this way. A handful of names lead everyone higher. The crowd piles in. The price runs miles ahead of the fundamentals. Then, one day, the crowd looks down, realizes how expensive it has gotten, and heads for the exit together.

The drop is fast, and it feels like the end of the world. It is not. This is what an overdue correction looks like, not the end of the AI story. The demand is still real. What died last week was the idea that you could pay any price for it. There is a difference, and it is the whole difference.

The spark for all of this came from an unlikely place. A Chinese startup called Moonshot released a new artificial intelligence model and said it performs on par with the leading systems from OpenAI and Anthropic. In a market priced for a handful of American giants to own this technology forever, that claim landed like a rock through a window.

Here is why it mattered so much. The entire chip trade rests on one belief. That the giants will keep spending hundreds of billions of dollars on AI, and that only they can. If a small team in China can match them for a fraction of the cost, that belief cracks, and so does the case for paying any price for the picks and shovels.

Dean’s note:
Let me be honest with you, because that is the whole point of this letter. I do not know if the Chinese model is really as good as they say. Neither does the market. Nobody has had time to find out. But that is exactly why it matters, and it is the signal to watch.

For two years, the story was that only a few giants could build this, and that they could spend whatever they wanted forever. One headline put a crack in that story. The fear is real, even if the model is not. So watch the big spenders. If they blink on their spending, the chips have a lot further to fall. If they double down, this scare fades. That single question is the ballgame now.

In the middle of the worst week for chips in over a year, the most important chipmaker on the planet reported the best quarter in its history. TSMC, the company that actually manufactures the advanced chips for Nvidia, Apple, and nearly everyone else, posted revenue up 36 percent and profit up more than 70 percent from a year ago. Demand, it said, is enormous.

And the stock fell anyway, right along with the rest of the group. A record quarter, a blowout number, real proof that the demand is not slowing, and the market shrugged and kept selling. If that seems backward, it is telling you something important about what actually happened last week.

Dean’s note:
This is the tell that this was a price problem, not a demand problem, and it is the most important thing in this whole issue. The company that makes the world's chips just had its best quarter ever, and its stock fell into a bear market the same week.

When great news cannot lift a stock, the market is telling you the good news was already baked into the price, and then some. The business is fine. The business is better than fine. It was the price that got sick, not the company. Do not confuse a correction in the stock for a collapse in the business. The picks and shovels are still flying off the shelf. The market just decided, all at once, to stop paying any price to own the store.

Lost in the chip drama was one of the most brutal single days a big American company has had in years. IBM, one of the oldest names in technology, fell about 26 percent in a single session after a rare profit warning. It was the worst day in the company's history, and it erased roughly $ 69 billion in value in an afternoon.

There was no missile involved, no Fed decision, no inflation shock. Just a company quietly admitting that demand in its software and infrastructure business was softer than promised. That was enough to take a quarter of the whole thing off the board before dinner.

Dean’s note:
Here is the lesson, and it is one you feel in your stomach, not your head. On a day when the inflation news was good, the banks were strong, and the headlines were mostly calm, a blue chip lost a quarter of its value in a few hours. That is the risk that hides under a smooth index.

A quiet market is not a safe market for every name inside it. This is why I keep preaching the boring gospel of diversification. It is the seatbelt you forget you are wearing right up until the day the car stops hard, and it saves your life. Friday was that day for anyone who had bet the farm on IBM, granted, not too many would. Never let one logo, no matter how old and trusted, become your whole story.

While everyone stared at the chips, the other half of the story kept building. The war got worse. The US reimposed its naval blockade on Iran's ports, Iran's Revolutionary Guard said it struck two more supertankers moving through the Strait of Hormuz, and crude surged more than 14 percent on the week. WTI ran past $82, and Brent cleared $88.

That is a long way from three weeks ago, when oil was on sale, and the only worry was a glut. The premium that drained out has come roaring back, and ship traffic through the strait has dropped sharply, even as millions of barrels still manage to squeeze through each day.

Dean’s note:
I keep coming back to oil because it keeps coming back to us, and this week it matters more than usual. Here is why. A chip correction, the market can handle on its own. A chip correction plus an oil shock is how a correction turns into something worse.

Cheap oil was the quiet hero of the first half of the year. It pulled inflation and yields down and let the Fed sit still. Now it has reversed, and it feeds straight into the inflation the Fed is watching and the costs every company pays. Remember, the cool June inflation number was built on cheap June oil, and that cheap oil is gone. So watch the barrel. It is the single line between a healthy correction and a real problem.

Stay invested. Stay selective. And do not confuse a correction in the most crowded trade with the end of the story.

A week ago, the AI trade finally cracked into a bear market, oil surged amid the war, and one blue chip lost a quarter of its value in a single day. Here is what I am holding onto.

•  The chips fell into a bear market, and that is mostly overdue, not alarming. The group had run up more than 100 percent. A trade that stretched that far finally snapped back. A great company and a great stock are not the same thing.

•  This was a price problem, not a demand problem. TSMC posted a record profit the same week the group fell 20 percent from its peak. The demand for chips is real. The market just stopped paying any price for it.

• The Chinese model is the story to watch. A startup called Moonshot claims it matches the best in the world. Whether it truly does or not, it made investors question the trillion-dollar spending race. If the big spenders blink, the chips have further to fall.

•  Oil is the wildcard. WTI ran past $82 and Brent near $88, up more than 14 percent on the week as the war escalated. A chip correction the market can digest. A chip correction plus an oil shock is a different animal. Watch the barrel.

•  The money rotated; it did not all run. The broad market held up far better than the Nasdaq. This looks like the crowded trade correcting while the rest of the market holds. Rotation, not ruin, at least so far.

•  One name can still get wrecked. IBM lost a quarter of its value in a day on a soft forecast. Under a calm index, a single stock can still blow up. Diversification is the seatbelt you forget you are wearing.

•  Keep contributing to your 401(k). The limit is 24,500 dollars, and if you are 60 to 63, you get a super catch-up to 35,750. A week when the hottest trade goes on sale is exactly when your automatic contribution quietly buys the dip for you, without you having to feel brave about it.

Here is where I land. This is the correction I have been telling you was coming in the most crowded corner of the market, and it finally arrived. So far, it is a healthy one, a stretched trade letting out air while the boring names hold their ground. The real danger is not the chips. It is the oil, because a spike there could turn a normal correction into something that spreads. Stay invested because demand is real and money is rotating, not fleeing. Stay selective, because the easy money in the crowded names is gone. And keep your eyes on the barrel, because that is the story that decides the next month.

- Dean

P.S. The number that sticks with me this week is 20. That is how far the chip index fell from its peak, in percent, officially a bear market. Now hold that next to the other number. That same group was up more than 100 percent on the way in. Up 105, then down 20. Sit with that. That is not a broken story. That is a wildly overbought one, letting air out. The companies still make the chips the whole world is desperate for, and TSMC just proved it with a record quarter. The price got miles ahead of the business, and now it's catching up. Up 105 then down 20 still leaves you far, far ahead. Do not confuse a correction for a collapse.

And one more thought. The next two weeks decide the mood, so brace for them. The biggest names in technology, the same giants that have promised to spend hundreds of billions of dollars on AI, report their earnings over the next fortnight. That is what the chip market is really waiting on. This whole selloff rests on a single question. Is the AI spending spree still on the way, or are the giants about to pull back? If they report and reaffirm the spending, the chips are likely to have found a bottom somewhere near here. If even one of them hints at tapping the brakes, this bear market has further to run. Stack the oil spike on top of all that, and the next two weeks are loaded. Do not trade the guess. Let the giants report. Their spending plans will tell you more than any headline about a model out of China.

👉 What if building real wealth in 2026 is not about calling the top in the AI trade or the next move in oil, but about owning a plan that already lived through a 100 percent chip rally, a war, and now a chip bear market, without ever betting everything on one of them? My Single-Digit Millionaire portfolio shows how to blend stocks, cash, gold, and even a little bitcoin so you are steady when the hottest trade cracks and positioned for whatever comes next. Take a look and see why this balanced approach can quietly put you ahead while everyone else is reacting to the noise.

This newsletter is for informational purposes only and does not constitute investment advice. Past performance is not indicative of future results. Consult with a qualified financial advisor before making any investment decisions.