What a week it's been.

The damage across the AI trade has been significant. Some of the biggest winners of this bull market have been absolutely crushed, and many charts look worse than they have in a long time.

And as if the market didn't have enough to worry about already, we now have another controversy surrounding Nvidia CEO Jensen Huang.

He had a relationship with a business executive suspected by US intelligence of working for China. The woman, Alice Huang, reportedly had connections to a company that was already under scrutiny over potential violations of US restrictions on advanced AI chip exports.

Jensen has since acknowledged the relationship, and Nvidia's board has launched an independent investigation into whether he was targeted by a foreign intelligence operation.

So far, there's no public evidence that Huang knowingly shared sensitive information with Chinese intelligence, and Nvidia says it has no reason to believe company secrets or national security were compromised.

It's another layer of uncertainty for a company that has been at the center of the AI trade and the broader market rally.

Does all of this mean we've bottomed?

I don't know, and I'm not particularly interested in trying to predict the exact bottom.

The technical damage is real, and there are still plenty of reasons to remain cautious. But after the amount of selling we've already experienced, I do think the risk/reward is becoming increasingly attractive in certain parts of the market.

I also want to repeat something I've been discussing throughout this report and over the past few weeks.

I don't believe the AI trade is dead. I believe it's changing.

The next generation of market leaders will probably look different from the companies that led the previous rally.

Every major market trend goes through different phases. Companies that led the first part of a bull market don't necessarily lead the next one. New businesses emerge, earnings expectations change, and investors start looking for opportunities elsewhere.

I expect that to happen with AI as well.

Just how bad was Thursday's selloff?

The S&P 500 fell 4.35% in a single day, making it the 124th worst trading day since 1928.

To put that into perspective, there have been more than 24,000 trading days over the past 100 years, and only 123 of them were worse. That's less than 1% of all trading days. And remember, this comes after weeks of selling, with many tech and AI stocks already down 30%, 40%, or even 50% from their highs.

Some of the biggest daily drops in history have happened closer to market bottoms than tops.

The S&P 500 has fallen for 4 weeks in a row and closed below all its major moving average for the first time since March, except for one. So, the good news is that SPY is still above its 200-day moving average, which has been the key support level during this bull market. For now, the bigger picture still looks decent. And after 4 weeks of losses, it's getting harder to ignore how much selling we've already seen.

The Nasdaq looks much worse in comparison.

But take a look at what happened earlier this year.

The last time we fell below the 200-day moving average was in March, and from there the market quickly bounced back and eventually reached new highs. Things looked pretty bad back then too. It’s worth remembering that markets can turn around quickly, especially when there's so much uncertainty.

But bottoms rarely form when everyone is feeling good about the market. They often happen when fear is high, stocks have been beaten down, and nobody wants to buy.

However, nothing good ever happens below the 200-day. So, this is the most important level to be reclaimed for any hope of a sustained rally.

This is already the 3rd time the S&P 500 has fallen more than 5% this year.

From January to March, the S&P 500 dropped 9.8%. Then in June, it fell another 5%. This time, the S&P 500 is down 6.3% from its recent high.

To put that into perspective, the typical pullback since 2009 has been around 7.6%.

So while the recent selling has been painful, we're still within the range of a normal market correction.

So far, this has been a pretty normal correction for the overall market. Since 1928, the S&P 500 has seen more than 3 pullbacks of 5% or more per year on average. And a bigger correction of 10% happens about once a year.

So far in 2026, we've had 3 drops of at least 5%, which is right in line with what we've seen in the past. Of course, that’s only true for the general market. Many growth and AI stocks have been hit much harder, with some already down 30%, 40%, or even 50% from their highs.

On Thursday, the VIX jumped to 28. That's its highest closing level since March 30 and its biggest one-day jump since Liberation Day in April 2025. Big jumps like these in the VIX tend to happen towards the end major market selloffs, when fear is already high.

The VIX jumped nearly 70%, its 3rd biggest one-day percentage increase since 1990.

What happened after similar spikes in the past?

Looking at the 20 biggest one-day percentage increases in the VIX since 1990, the S&P 500 went on to gain an average of:

• 1 month later: +2%

• 3 months later: +5%

• 6 months later: +10%

• 1 year later: +17%

Thursday's selloff didn't come out of nowhere. Tech stocks have been struggling for a while. Just 1 month ago, the tech sector was up 31% for the year. Now it's up just 9.5%.

Meanwhile, sectors like health care, consumer staples, and materials are catching up after falling behind for most of the year. Investors have been moving money out of tech for weeks. That's also why the S&P 500 has held up relatively well, even as many tech stocks have fallen sharply.

In fact, the S&P 500 excluding AI stocks even reached a new all-time high.

So, investors are moving money from one sector to another rather than leaving the stock market altogether.

The AI trade was already under pressure, and now investors have another reason to worry.

Following the Democrats' win in the US midterm elections, investors are concered that stricter regulations could make it more difficult to build new AI infrastructure, particularly data centers.

And the election results show just how much opposition there is.

Across the US, voters supported restrictions on data center development in 40 out of 41 local ballot measures. The only exception was a small town in North Dakota.

For the AI trade, this creates another challenge. Technology companies are planning to spend hundreds of billions of dollars expanding their AI infrastructure, and tougher regulations could make those projects more expensive, delay construction, or force companies to look for alternative locations.

And given how much of the bull market has been driven by expectations of continued AI spending, it's easy to see why investors are cautious.

The bigger concern is that much of the planned AI infrastructure hasn't even been built yet. Companies have announced massive expansion plans, but stricter regulations make it harder and more expensive to turn those plans into reality.

Some AI infrastructure companies are particularly exposed to this risk.

Take IREN, for example. Nearly half of its announced power capacity is located in Texas, while CoreWeave has several data center sites in Texas and Ohio, where opposition to new developments is growing.

There were already concerns about excessive AI spending, stretched valuations, and whether companies will ever generate enough revenue to justify their massive investments.

Now investors are also questioning whether these companies will actually be able to build all the infrastructure they've promised.

That adds another layer of uncertainty to an AI trade that's already under significant pressure.

I still believe the AI trade has a lot of room to grow. But as I’ve said several times, I also believe the next generation of market leaders will look very different from the companies that led the previous rally.

Every stock goes through a cycle, and one of the biggest mistakes investors make is assuming that the stocks that led the last rally will automatically lead the next one.

Some will eventually recover and make new highs. Others might spend months moving sideways or underperforming while new leaders start to emerge elsewhere.

And I think that's exactly what we need to start paying attention to.

I believe the next big winners of the AI trade will increasingly come from companies that find ways to turn AI into higher revenue, better profit margins, and stronger earnings.

We've already seen enormous amounts of money flow into the infrastructure needed to build and run AI. Now I expect investors to start paying much closer attention to the companies that can actually use that technology to improve their businesses.

We're already seeing early signs of this shift in cybersecurity and agentic AI, and I think it's only a matter of time before it spreads into other industries.

A great example came from Airbnb's latest earnings report.

CEO Brian Chesky described AI as one of the best things to ever happen to Airbnb, and the results are starting to show why.

In Q3, around 40% of all customer support issues were resolved without a human agent, up from roughly 33% in the previous quarter. This helped Airbnb reduce its customer support costs per booking by more than 17% compared to a year ago.

And the benefits go beyond customer support.

Airbnb reported that it launched more than 50 new products and features during the quarter, nearly 70% more than a year earlier, with AI helping its teams develop and release products faster.

Airbnb is using AI to serve customers more efficiently, lower operating costs, and develop new features faster. These improvements can eventually translate into higher profit margins and stronger free cash flow. And Airbnb is doing all of this without having to build massive data centers or manufacture GPUs.

It's using AI to make an already successful business even better.

That's the kind of opportunity I expect to see across more companies, particularly in consumer-facing businesses that already have millions of customers and established distribution.

They don't necessarily need to create the next breakthrough AI model. They can use the technology that's already available to make their products better, increase productivity, and grow earnings without having to increase costs at the same pace.

Of course, using AI alone won't make a company a great investment. Investors will want to see those improvements show up in the financial results.

The first thing I want to see is the Nasdaq reclaim its 200-day moving average. A quick move back above that level would be an encouraging sign that Thursday's selloff was just a temporary shakeout. The longer it stays below, the greater the risk that we see more sideways trading and volatility in the weeks ahead.

The S&P 500 is still holding above its own 200-day moving average, and I'd like to see that continue.

Looking at the bigger picture, this correction was needed.

The bull market has had an incredible run, and the AI trade in particular had become extremely extended. Expectations were high, valuations were stretched, and investors were willing to pay almost any price for companies connected to AI.

A period of consolidation gives the market a chance to cool down, allows valuations to become more reasonable, and creates a healthier foundation for the next move higher.

And when it comes to AI, I think it's important to remember what hasn't changed.

The underlying demand for AI infrastructure remains enormous. Companies are still investing billions into building data centers, expanding computing capacity, and developing more advanced AI models.

That demand doesn't suddenly disappear because Democrats won a few elections or because some states are becoming more restrictive about data center construction.

The competition between the world's largest technology companies is still intense, and none of them wants to fall behind.

If anything, the pressure to keep investing remains incredibly high.

And the United States isn't the only place where companies can build.

Countries such as Saudi Arabia, the UAE, and Canada are actively trying to attract AI infrastructure investment by offering access to energy, land, and potentially faster approval processes.

If certain states make it too difficult to build new data centers, companies will look elsewhere. And as local governments start weighing the jobs, tax revenue, and economic activity these projects can bring, some may eventually reconsider their positions.

Across the country, there are underused industrial buildings and other facilities that could potentially be converted into data centers, although access to electricity, cooling, and network connections remains a major challenge.

I expect we'll see more creative solutions as companies look for ways to expand capacity without waiting years for entirely new projects.

And we've been through similar scares before.

Remember the panic around Liberation Day and Trump's tariff announcements in April 2025? Or when DeepSeek raised concerns that AI models might become much cheaper to develop, putting the entire AI infrastructure trade into question?

Both events triggered significant selling and made investors question whether the AI boom had finally peaked.

Yet the underlying trend continued, and many AI-related stocks eventually recovered and moved to new highs.

I think it's worth keeping that perspective in mind during the current correction.

Looking across the charts, we're seeing some extreme selling pressure and significant declines in valuations, particularly among the stocks that have led the AI trade.

Hedge funds have been selling AI-related stocks at their fastest pace in more than 5 years. The VIX recently spiked by around 70%, the put/call ratio has climbed to its highest level since the April 2025 market bottom, and bearish sentiment is approaching its highest reading in a year.

These are significant moves, especially when they happen together.

What's interesting is that despite all this damage beneath the surface, the S&P 500 is still only around 6% below its recent high.

From a historical perspective, that's a fairly normal pullback. But it certainly doesn't feel normal when you look at how badly many individual stocks have been hit.

Some of the biggest leaders of this bull market have already suffered massive declines, with many falling much further than the broader indexes.

And after weeks of selling, I'm finally starting to notice a change in sentiment.

For much of this correction, investors seemed convinced that every dip was another opportunity to buy and that the market would eventually bounce right back.

I'm seeing more people calling for the end of the bull market, with comparisons to the dot-com bubble in 2000 and even the financial crisis in 2008 becoming increasingly common.

The optimism that seemed almost impossible to shake just a few weeks ago is disappearing.

For the first time during this correction, it feels like we're starting to see investors genuinely lose confidence.

At some point, expectations become so low that it doesn't take much good news to change the direction of the market.

But there's no reason to rush.

The biggest mistake investors can make in an environment like this is allowing fear, frustration, or FOMO to dictate their decisions.

We've had an incredible run in many parts of the market, and there's nothing wrong with waiting for better conditions.

Some of the best opportunities emerge after major drawdowns and periods of extreme volatility.

We're experiencing both right now.

I don't know whether the market has another 5% or 10% downside ahead, or whether we're much closer to the bottom than people realize.

What I do know is that the longer-term opportunity in AI remains substantial, and periods like this have historically created attractive entry points for investors who were patient enough to wait for them.

The key is making sure we still have the capital and confidence to take advantage of those opportunities when they arrive.

Fear creates opportunity. But patience is what allows you to take advantage of it.