The chip trade is back.

Semiconductors roared back into leadership this week, helping push the NASDAQ to a new high as the market moved toward Friday with renewed optimism surrounding the U.S.-China discussions.

The headline is familiar: AI, semiconductors, and technology are leading again.

But underneath the surface, the market is telling us something very different.

Market breadth remains extremely narrow.

While technology has surged, a large number of sectors have retreated significantly from their recent highs. Industrials, consumer discretionary, consumer staples, real estate, financials, basic materials, and utilities have all experienced meaningful pullbacks.

This creates an unusual market structure:

The index is making new highs, while a large part of the market remains well below its highs.

That divergence deserves attention.

It may simply represent continued concentration in the AI trade.

Or it may be the early stage of something more interesting:

a sector rotation waiting to broaden.

The answer appears to be yes.

Semiconductors have regained momentum and are once again pulling the NASDAQ higher.

The fundamental narrative remains powerful.

AI infrastructure continues to require enormous amounts of computing power, networking, memory, data-center capacity, and electricity. The semiconductor industry sits directly at the center of that investment cycle.

But there is an important distinction between AI infrastructure growth and the next phase of AI economic adoption.

The first phase was dominated by hyperscalers and the infrastructure required to build increasingly powerful AI systems.

The next phase could increasingly move toward the application layer.

And that transition could look very different in the stock market.

The first phase of the AI boom was largely a large-cap technology story.

NVIDIA, Broadcom, Microsoft, Amazon, Alphabet, Meta, and other major technology companies have been the primary beneficiaries of the massive capital expenditure cycle.

But as AI infrastructure becomes increasingly established, the economic value may begin moving down the stack.

The next question is no longer simply:

Who is building the AI infrastructure?

It is:

Who is actually using AI to transform their business?

That is where the application layer becomes increasingly important.

And this is where we believe the opportunity could broaden into areas of the market that have not participated nearly as much in the current rally.

Healthcare.

Genomics.

Financial services.

Enterprise software.

Industrial automation.

Smaller technology companies.

The AI revolution may ultimately become less about the handful of companies building the largest models and more about the thousands of companies figuring out how to use those models.

One of the most interesting consequences of this transition could be the expansion of AI into the small- and mid-cap universe.

Large technology companies have the capital, data, infrastructure, and talent to build AI.

Smaller companies have a different potential advantage:

They can use AI to fundamentally change their cost structure.

A company does not need to build a frontier model to benefit from AI.

It may simply need to deploy one effectively.

Imagine a smaller financial-services company using AI to automate underwriting, compliance, research, customer service, or fraud detection.

Imagine a healthcare company using AI to accelerate drug discovery or analyze genomic data.

Imagine a software company replacing labor-intensive workflows with AI agents.

The economic impact could be substantial even if the underlying AI model is built by someone else.

This is why we believe the next phase of AI growth may increasingly show up in earnings statements rather than capital-expenditure announcements.

Healthcare and genomics remain particularly interesting in this framework.

The combination of AI, biological data, computing power, and increasingly sophisticated modeling could transform drug discovery, diagnostics, clinical trials, and personalized medicine.

The sector also has an important characteristic that distinguishes it from the current semiconductor trade:

Its AI opportunity is still relatively early.

The market has already spent years pricing the AI infrastructure buildout.

The application of AI to biology and healthcare is much less mature.

That does not mean every healthcare or genomics company will benefit.

Biotech remains a high-risk industry with clinical, regulatory, financing, and commercialization risks.

But from a thematic perspective, the intersection of AI + biology + genomics could become one of the more important technology-driven growth stories of the next decade.

This is where the sector rotation becomes particularly interesting.

Financials have been left behind while technology has dominated.

The XLF has declined approximately 3.7% over August and September, while technology has gained nearly 12%.

That divergence is significant.

Financial services could become an important beneficiary of the next stage of AI adoption.

Think about the number of processes within financial institutions that involve:

These are precisely the types of workflows where AI can potentially create substantial productivity gains.

And unlike the semiconductor industry, financial institutions do not need to spend hundreds of billions of dollars building AI infrastructure.

They can consume AI infrastructure and potentially convert it into higher productivity and lower operating expenses.

That creates a very different economic model.

The market has spent the last several years rewarding companies that sell the picks and shovels of AI.

The next phase may increasingly reward companies that use the picks and shovels effectively.

Financial services could be one of those areas.

Here is the performance picture for August and September 2026:

The numbers tell an interesting story.

Technology has separated itself dramatically from the rest of the market.

Healthcare, communication services, and energy have also remained positive.

But almost everything else has struggled.

The weakest areas—utilities, real estate, industrials, consumer discretionary, financials, and consumer staples—have experienced substantial relative deterioration.

This is what makes the current market structure so unusual.

The NASDAQ is making new highs, but the average sector is not participating.

This is the question we are watching most closely.

When leadership becomes extremely concentrated, there are generally two possibilities.

The leading sector can continue to dominate while the rest of the market remains weak.

Or the market can begin to broaden, with capital rotating into sectors that have already experienced substantial corrections.

We are beginning to see some signs that the latter could be developing.

That does not mean the technology trade is finished.

Quite the opposite.

Technology is leading the market because the AI investment cycle remains powerful.

But leadership and opportunity are not necessarily the same thing.

Once a sector has already experienced a major move, investors naturally begin looking for the next areas where earnings growth, valuation, and improving momentum can intersect.

That is where the beaten-down sectors become interesting.

Chips are roaring back.

We maintain exposure to semiconductors because the AI infrastructure cycle remains one of the most powerful structural growth trends in the market.

But precisely because semiconductors are once again leading, it may be time to start looking beyond the obvious winners.

The market does not necessarily need to abandon technology for the next opportunity to emerge.

It simply needs to broaden.

And there are plenty of sectors where expectations have already been reset.

Financial services stand out.

The sector has underperformed technology substantially during August and September, yet the underlying AI opportunity is only beginning to develop.

Healthcare and genomics are another area we are watching closely.

The AI application layer could create an entirely new wave of productivity and innovation in these industries.

And this may ultimately be the more interesting part of the AI story.

The first wave of AI created infrastructure.

The second wave may create productivity.

That distinction could have major implications for sector allocation.

The current market structure can be summarized in one sentence:

The index is strong, but breadth is weak.

That is not necessarily bearish.

It may simply mean that the market is still concentrated in the strongest earnings-growth story available—AI and technology.

But if breadth begins to improve, the impact could be significant.

Consider what happens if capital starts rotating into sectors that have already experienced substantial corrections.

Financials.

Industrials.

Consumer discretionary.

Materials.

Real estate.

Even utilities.

A small improvement in sentiment combined with improving earnings expectations could create a meaningful momentum shift because positioning in many of these areas is already considerably less extended than in technology.

This is where we will be watching for bottoming patterns and improving relative momentum.

The key signal will not simply be that these sectors stop falling.

It will be whether they begin to outperform the market after periods of under-performance.

The market has spent enormous amounts of capital building the infrastructure required for AI.

Now comes the difficult part.

Make money from it.

That means deploying AI into actual businesses.

Healthcare companies need to use AI to discover drugs faster.

Financial institutions need to reduce operating costs.

Software companies need to automate workflows.

Industrial companies need to improve productivity.

Small businesses need to operate with fewer people.

The winners of the next phase may therefore look very different from the winners of the first phase.

They may not be the companies selling the most GPUs.

They may be the companies that demonstrate the largest improvement in:

That is the application layer we are watching.

The chip trade is back.

The NASDAQ is making new highs.

But beneath the surface, the market is becoming increasingly narrow.

Technology has dominated the August–September rotation, while several major sectors have experienced corrections of 5%, 8%, and even 10% or more.

That creates an interesting setup.

We are not suggesting abandoning the AI trade.

Instead, we are looking for the next baton pass.

The first baton was infrastructure.

The next may be application.

And if that transition begins to show up in earnings, healthcare, genomics, financial services, and selected smaller technology companies could become increasingly important parts of the AI story.

At the same time, the beaten-down sectors deserve attention.

A sector does not need to be the current leader to become the next leader.

The question is whether the market is beginning to build a base.

As technology leads the renewed rotation, we will be watching closely for signs that capital is beginning to move into the sectors that have been left behind.

The chip trade may be back.

But the more interesting question may be:

What comes after the chips?