To The Moon

QQQ closed Friday at:

$690.49

down:

3.47%.

The Nasdaq fell almost 3%.

Semiconductors were crushed.

AMD fell 10.9%.

Nvidia fell 4.2%.

Intel collapsed 28.6%.

And suddenly the question everywhere is:

IS THE AI TRADE FINALLY BREAKING?

I don't think that is what Friday told us.

I think something more important happened.

For two years, the market rewarded companies for:

SPENDING ON AI.

Now it is beginning to ask:

WHO ACTUALLY EARNS A RETURN FROM IT?

That is not necessarily the end of the AI bull market.

It may be:

THE END OF EASY AI MONEY.

And that distinction completely changes how I would buy this dip.

🔥 FRIDAY WAS NOT A NORMAL RISK-OFF DAY

Look underneath the indices.

QQQ:

-3.47%

S&P 500:

-1.82%

Dow:

-1.09%

Nasdaq:

-2.86%

That dispersion matters.

If investors suddenly believed the entire US economy was falling apart, I would expect much broader destruction.

Instead?

The epicentre was:

TECH + AI + SEMICONDUCTORS.

That tells me Friday looked less like:

“SELL EVERYTHING”

and more like:

“REPRICE THE AI TRADE.”

That is a very different signal.

🧠 WHAT ACTUALLY BROKE?

Not AI demand.

Not yet.

What broke was the assumption that:

ALL AI CAPEX DESERVES THE SAME VALUATION.

Look at what earnings season just taught us.

Microsoft showed enormous AI spending AND positive free cash flow.

The market can tolerate that.

Google showed extraordinary Cloud growth but negative free cash flow after enormous capex.

The market punished it.

Amazon showed accelerating AWS growth but massive investment and deteriorating cash generation.

Investors became nervous.

Meta is spending aggressively too.

Then Intel delivered weak guidance.

Suddenly the market connected the dots.

The question changed from:

“HOW MUCH ARE YOU SPENDING ON AI?”

to:

“WHEN DO I GET MY MONEY BACK?”

That is the watershed.

🔥 I THINK AI HAS ENTERED ITS FOURTH PHASE

We have been building toward this across my recent articles.

PHASE 1

WHO HAS AI?

ChatGPT.

Gemini.

Claude.

Copilot.

PHASE 2

WHO CAN BUILD AI?

GPUs.

HBM.

Data centres.

Networking.

Power.

PHASE 3

WHO CAN MONETISE AI?

Azure.

Google Cloud.

AWS.

Advertising.

Enterprise software.

AND NOW PHASE 4:

WHO CAN EARN AN ACCEPTABLE RETURN ON AI CAPITAL?

This is the phase where:

revenue isn't enough

growth isn't enough

orders aren't enough

capex isn't automatically bullish.

Investors want:

CASH.

That makes the next stage of the AI trade much more selective.

🔵 SO SHOULD WE BUY QQQ AT $690?

I would.

But:

NOT ALL AT ONCE.

This is where I think people make one of two mistakes.

One camp sees a 3.5% crash and panics.

The other sees “November seasonality” and immediately goes all-in.

I would do neither.

QQQ is now meaningfully below its October high.

That improves the risk/reward.

But a 3.5% daily decline does NOT automatically mean:

BOTTOM.

The Nasdaq is still digesting an enormous valuation reset in AI.

And once markets begin questioning capital returns, one trading day rarely answers everything.

So I would:

SCALE INTO FEAR.

Not predict the exact bottom.

📍 MY QQQ BUYING MAP

Around:

$690

I would treat QQQ as:

🟡 STARTER POSITION

Not maximum exposure.

If QQQ stabilises and reclaims:

$700-$705

I would want to see whether buyers can defend that recovery.

That would be my first indication Friday's liquidation is being absorbed.

If we fall toward:

$665-$680

🟢 SECOND BUY

This is where I would become more aggressive.

A further 2%-4% decline would remove considerably more excess without requiring the AI thesis to break.

If panic takes QQQ toward:

$635-$655

🟢 PREFERRED BUY ZONE

Assuming:

AI demand remains intact

hyperscaler capex remains intact

earnings estimates aren't collapsing

I would want significantly more exposure.

But if QQQ breaks:

BELOW $630

I stop automatically calling it a dip.

🔴 THESIS CHECK

At that point I want to know whether the market is pricing:

recession

widespread earnings revisions

actual AI capex cuts

or:

a deeper valuation compression.

The lower the price goes, the more important the WHY becomes.

🚨 BUT I DON'T THINK QQQ IS THE BEST WAY TO PLAY THE DIP

This is where my view becomes more selective.

QQQ gives you everything.

The companies proving AI economics.

AND:

The companies still burning cash trying to prove them.

Why buy both equally?

After this earnings season, I increasingly want to separate AI into:

🟢 THE CASH GENERATORS

Companies already showing that AI can generate revenue AND cash.

My clearest example:

MSFT

Azure +48%.

Operating cash flow $58B.

FCF $18.6B despite $45.2B capex.

That is what the market increasingly wants.

🟢 THE BOTTLENECK SUPPLIERS

These companies don't need to determine which hyperscaler eventually wins.

They get paid while everyone builds.

NVDA

GPUs.

MU

Memory.

SNDK

Storage.

Microsoft can win.

Google can win.

Amazon can win.

Meta can win.

The suppliers can potentially sell to:

ALL OF THEM.

That is why I still like this layer of the AI stack.

🟡 THE AI BUILDERS

Alphabet.

Amazon.

Meta.

These companies may ultimately earn extraordinary returns from today's investments.

But now?

The burden of proof has increased.

The market wants evidence that:

CAPEX → REVENUE → MARGIN → CASH.

Until that becomes clearer, volatility stays high.

🔴 THE AI STORY STOCKS

This is where I would be most careful buying the dip.

Companies whose valuation depends on:

future AI revenue

future customers

future margins

future cash flow.

In the first stage of an AI bull market, stories can outperform fundamentals.

In this stage?

THE MARKET IS STARTING TO SEND INVOICES.

🔥 HERE IS THE PARADOX EVERYONE SHOULD UNDERSTAND

Alphabet shareholders may hate $51B quarterly capex.

Amazon shareholders may hate enormous infrastructure spending.

Meta shareholders may hate higher capex guidance.

But somebody receives that money.

Think about the chain.

Hyperscaler:

CAPEX OUT

↓

AI supplier:

REVENUE IN

That is why I am not automatically bearish on:

NVDA

MU

SNDK

simply because hyperscaler free cash flow is under pressure.

In fact?

The very thing hurting hyperscaler cash flow can strengthen supplier revenue.

The real danger for suppliers begins when hyperscalers say:

“WE ARE CUTTING AI CAPEX.”

Not:

“AI CAPEX IS EXPENSIVE.”

Those are completely different statements.

And so far?

The major builders are still spending enormous amounts.

🧠 THIS ALSO CHANGES HOW I READ AMD

AMD fell almost:

11%

Friday.

That looks terrifying.

But remember the question from my AMD earnings article.

AMD already has the AI orders.

What it needs to prove is:

ORDERS → SHIPMENTS → REVENUE → MARGIN → CASH.

Friday's selloff actually makes AMD's upcoming earnings MORE important.

Because the market just announced what it wants.

Not another giant customer.

Not another gigawatt announcement.

It wants:

AI ECONOMICS.

If AMD delivers that?

The selloff could create opportunity.

If it doesn't?

A cheap-looking stock can become cheaper very quickly.

🔥 WHAT ABOUT NOVEMBER SEASONALITY?

This is the part I would use carefully.

Historically, November has been one of QQQ's stronger months.

Across QQQ's history, November has averaged roughly:

+2.6%

and finished positive about:

74% OF THE TIME.

Interesting.

But seasonality is:

A TAILWIND.

Not:

A THESIS.

November cannot rescue collapsing earnings.

November cannot rescue falling AI capex.

November cannot rescue a recession.

But if:

earnings remain healthy

AI demand remains strong

capex remains intact

Friday was mainly valuation compression

then seasonality becomes useful.

Because now we have:

LOWER PRICES + STRONGER SEASONAL WINDOW.

That is much more interesting than buying QQQ simply because the calendar changed.

🔮 MY THREE QQQ SCENARIOS

🟢 BULL CASE: 30%

Friday was largely forced de-risking and valuation compression.

AI demand remains strong.

Apple/AMD and later Nvidia reassure investors.

QQQ quickly reclaims:

$705

Then:

$725-$740

becomes possible.

And eventually the October high comes back into play.

🟡 BASE CASE: 50%

QQQ doesn't immediately V-shape.

Instead we get:

violent rallies

renewed selling

earnings-driven rotations

while the market reprices individual AI names.

My expected range:

$665-$715

before a clearer direction emerges.

This is actually my preferred environment.

Why?

Because volatility creates:

BETTER ENTRIES.

🔴 BEAR CASE: 20%

The AI capex concern becomes an actual earnings problem.

Hyperscalers begin cutting spending.

AI suppliers lower expectations.

Earnings estimates fall.

Then:

$635-$655

may not be the bottom.

And below:

$630

I reassess the broader thesis.

🚨 WHAT WOULD MAKE ME STOP BUYING THE DIP?

This is probably more important than predicting Monday.

I would become significantly more defensive if I see:

🔴 hyperscalers CUT AI capex

not merely complain about its cost.

🔴 cloud growth decelerate sharply

especially Azure, AWS and Google Cloud.

🔴 memory pricing weaken despite AI demand.

That would challenge my MU/SNDK thesis.

🔴 Nvidia demand commentary weaken.

That would challenge the infrastructure thesis.

🔴 credit spreads widen sharply.

That would tell me the problem is moving beyond equity valuation.

🔴 earnings estimates start falling across sectors.

That would signal something broader than an AI reset.

Until those appear?

I treat this as:

A REPRICING EVENT

rather than:

THE DEATH OF THE AI CYCLE.

MY CALL

Would I buy QQQ after Friday?

YES.

Would I go all-in Monday?

NO.

My strategy is:

$690 area

START

$665-$680

ADD

$635-$655

BUY FEAR

provided the earnings thesis remains intact.

But there is an even bigger lesson here.

The AI trade isn't disappearing.

IT IS MATURING.

And mature markets become selective.

The winners of the first AI phase were companies that could say:

“WE HAVE AI.”

The winners of the second phase were companies that could say:

“WE ARE SPENDING ON AI.”

I think the next winners will be companies that can say:

“HERE IS WHAT WE EARNED FROM AI.”

That is why Friday did not make me abandon the AI trade.

It made me:

UPGRADE MY FILTER.

Less narrative.

More margins.

Less capex announcements.

More cash conversion.

Less:

“HOW BIG IS THE AI OPPORTUNITY?”

More:

“WHO GETS PAID?”

And after a 3.5% QQQ selloff?

That is exactly the question I want to be asking BEFORE I buy the dip.

I am not a financial advisor. Trade wisely, Comrades!

Are you buying QQQ itself after Friday, or using the selloff to pick individual AI winners instead?

If you find this post interesting, give it wings! 🪽 Repost and share the insights.

Do consider ‘Follow me’ and get firsthand read of my daily new post. Thank you.

@TigerPM  @TigerStars  @Daily_Discussion  @TigerEvents  @Tiger_SG