The main stock indices had a solid week, but it was a tougher ride for anyone focused on semiconductor stocks.

The S&P 500 gained 0.49%, the Nasdaq Composite rose 0.85%, and the Dow added 0.53% for the week ending 28 August. Friday told a different story. The S&P 500 lost 0.25%, the Nasdaq fell 0.52%, and the Dow slipped 0.02%.

The difference matters.

Earlier in the week, Nvidia reported another strong quarter. Revenue more than doubled to $96.22 billion, and management expects about 70% revenue growth for the fiscal year ending January 2028. The stock price jumped 8.7% at first.

Then came Friday.

Nvidia dropped 4.6%. Marvell fell 10.3% as investors wondered how soon its new partnership with Alphabet would boost revenue. Lam Research was down about 5.2%, and Applied Materials lost 4.3%.

Nothing particularly dramatic happened to the long-term demand for artificial intelligence during those few hours.

It just became harder to justify the high prices for those stocks.

At index level, the week looked calm:

But Friday’s semiconductor tape looked very different:

_{Sources: Reuters and market closing data, 28 August 2026.}

That was the most interesting part of the week.

The market did not reject AI.

Investors became more careful about which AI stocks they were willing to pay a premium for.

Nvidia’s quarterly revenue reached $96.22 billion, up more than 100% year on year, while management projected approximately 70% growth for fiscal 2028.

Demand remains broad across hyperscalers, enterprises and sovereign AI projects. Nvidia also said supply remains constrained, particularly as memory availability and component costs become increasingly important.

This difference is important.

The constraint facing AI infrastructure is gradually shifting from whether customers want more compute to whether the industry can physically supply everything required to build it.

That brings Micron, SK hynix, TSMC, Lam Research, Applied Materials, and ASML deeper into the story.

Federal Reserve Chair Kevin Warsh used his first Jackson Hole speech to reaffirm the Fed’s commitment to its 2% inflation objective.

Markets responded by increasing expectations for another rate increase. As of Monday morning, pricing implied roughly a 58% probability of a September hike.

That matters most for companies in which a large share of the valuation depends on profits expected many years from now.

AI may be changing computing.

But it hasn’t changed how discounted cash-flow calculations work.

The next major market events are unusually concentrated.

The August employment report arrives on 4 September, followed by PPI on 10 September, CPI on 11 September, and the Federal Reserve decision on 16 September.

Right now, the situation is uneasy but not disastrous.

July payrolls declined by 23,000, with unemployment at 4.1%. At the same time, headline PCE inflation reached 3.7%, with core PCE at 3.3%.

Weak employment usually argues against tighter policy.

Persistent inflation argues for it.

The Fed gets to choose which problem worries it more.

Seeing Nvidia drop 4% or Marvell fall 10% looks dramatic on a chart.

But Nvidia had risen 8.7% immediately following its results, while Marvell entered earnings after a remarkable run of its own. Large moves work in both directions when expectations are high.

The important question is not whether semiconductor shares fall after earnings.

The real question is whether the revenue forecasts supporting these stock prices begin to weaken.

Nvidia’s latest outlook suggested the opposite.

Warsh’s Jackson Hole speech mattered because it reinforced the inflation objective.

It did not settle the September decision.

The Fed receives the August employment report, PPI, and CPI before its 15 to 16 September meeting.

Markets can react to interest rate news instantly.

The Fed only has to make one on 16 September.

Marvell illustrates the difference.

The company delivered strong growth and raised longer-term expectations, but investors focused on the timing of material revenue from its custom-chip relationship with Google. Management indicated that the largest contribution is likely to arrive later than some investors had hoped.

The business outlook improved.

The difference between what investors expected and when it would happen was disappointing.

Both things can be true.

The clearest representation of last week is not a chart of the S&P 500 alone.

It is the contrast between the weekly index return and the Friday semiconductor sell-off.

This comparison uses weekly index returns and Friday’s stock moves on purpose to highlight the contrast.

The market finished the week higher, even though some of its most popular AI infrastructure names finished the week badly.

This isn’t a contradiction.

It’s an example of market rotation.

Very little changed in the underlying demand picture.

Nvidia’s outlook points to continued expansion in AI computing. Memory intensity is increasing. Hyperscalers are still committing enormous amounts of capital to infrastructure. Networking, storage, power, and semiconductor manufacturing capacity remain part of the same build-out.

What changed is the required return for these investments.

When bond yields rise, the market demands more compensation for owning long-duration growth assets.

This leads to a situation in which Nvidia can report strong results, but its stock still drops a couple of days later.

A great company can produce great earnings.

A good stock also needs to be priced so that strong earnings actually make a difference.

AI is moving from experiment… to essential.

Every major industry is integrating it.
Every major company is investing in it.

By late 2025, AI was already an $800B market — growing at a pace that could push it well beyond $1 trillion in the years ahead.

Cloud infrastructure is scaling fast.
AI-enabled devices are multiplying.
Automation is becoming standard.

But here’s the real question…

When trillions flow into this transformation — which stocks stand to benefit most?

Our new report reveals 10 AI stocks positioned across the backbone of this shift — from the companies powering the infrastructure… to those embedding intelligence into everyday systems.

If you want exposure to one of the defining growth trends of this decade, start here.

Download the Report Now

September begins with three competing forces: