October’s Warning Shot: Is AI Entering Its Dot-Com Danger Zone? November 6, 2026

OCTOBER LOOKED BAD. BUT WAS IT?

Welcome back to Stock Talk. October just gave investors one of the strangest months of this entire bull market. If you only owned the biggest AI, NASDAQ 100, MAG 7, large cap growth,  and technology names, October probably felt terrible.But here’s the surprise. Most S&P500 stocks actually went UP.

The S&P 500 fell – 0.9% in October. Technology fell -7%. Communication Services fell more than -15%. Thank you Google, META and wireless telco providers.

Yet nine of the eleven S&P sectors finished higher. Energy gained almost 14%.

Consumer Staples gained more than 11%.Utilities gained more than 10%. And the average S&P 500 stock? It gained almost 4.8%. In fact, 352 of the 500 stocks went up.That’s not a normal bear-market selloff.

That’s a rotation.And it raises a much bigger question. Is this simply money moving out of crowded AI winners and into the rest of the market?Or is AI starting to enter the dangerous phase we saw during the final months of the dot-com boom?Because underneath the market, something else is happening.

AI capital spending is exploding. Oil has exploded. Interest rates are high. Inflation is sticky. The Fed is tightening. And the cost of financing the AI boom is finally starting to matter. If this sounds familiar? It should. Because that is where our comparison with lathe late Dotcom cycle gets very interesting.

Today, we’re going to look at three things: October. 2026 year-to-date. And finally: 2026 AI versus the final innings of dot-com.

THE INDEX HID WHAT REALLY HAPPENED

Here’s the first thing I don’t want you to miss. October wasn’t really a story about the S&P 500 falling 0.9%.

It was a story about who owned what. Bespoke found that October was the equal-weight S&P 500’s best month relative to the regular S&P 500 since April 2009. Yes, since 2009, the trough of the GFC.

Franklin Templeton sees the same thing.Through October 28: Russell 1000 Value: +24.4%. Russell 1000 Growth: only +4.5%. Small-cap value: +20.3%. Small-cap growth: +10.7%. And the equal-weight S&P 500 was beating the regular cap-weighted index. That’s a major change in leadership. One that had been going on slowly for about 4-5 month, until last week when it exploded higher.

THE EARNINGS ARE STILL THERE

So point number one is: October was a leadership correction – not yet a market correction.

And earnings help explain why.This is where the bubble argument gets harder.

Q3 S&P 500 earnings are running around 30% above last year, depending on the methodology used. Goldman Sachs estimates about 33% growth. That’s extraordinary. But there’s a catch.

The growth will no longer be accelerating from here. Goldman expects roughly 30% EPS growth in 2026, slowing toward about 18% in 2027. That’s not an earnings recession. It’s a change in the rate of change.

And markets care about rate of change – especially when expectations are sky-high.That’s why companies could beat earnings in October and still get crushed.Goldman found Technology companies beating earnings generally were not rewarded like the broader market.That’s telling us: Good news is no longer good enough for the old leaders.

NOW LOOK AT THE WHOLE YEAR

Now step back. Because 2026 looks completely different when we stop staring only at Nvidia and AI.

Here’s Goldman’s year-to-date scoreboard. Think about that.

The financial story of 2026 isn’t simple: AI wins. It’s: Nominal growth wins.

Oil. Energy. Value. Healthcare. And yes – technology. That’s a much broader economic message.

THE ECONOMY IS STRONG, BUT EXPENSIVE

Real GDP grew 3.4% year-over-year in Q3, according to Franklin Templeton’s October market review. Their 2026 GDP forecast is about 2.5%. Let’s call the economy: Good, but not great. The labor markets are overstated and half of the growth in AI related. Manufacturing remains positive. Initial unemployment claims are around 200,000. Retail sales are growing more than 6%. Retail sales excluding autos are growing more than 7%. But housing is weak. Consumer confidence is weak. And inflation has become the problem again. September CPI: 3.72%. Core CPI: 2.57%. PCE inflation: 3.65%. Core PCE: 3.12%. Import prices: +8.1%. That doesn’t look like an economy heading into recession. But it also doesn’t give the Fed much room to cut rates.

THIS MAY BE THE MOST IMPORTANT CHART

The two-year Treasury is around 4.6%. The ten-year is around 5%. Both down 30-40bps in the last 4 weeks. The 2-year is above the Fed funds rate and the bond market is calling for another Fed hike. Fed funds futures were assigning roughly a 79% probability of another quarter-point hike in December. And Bespoke notes the 10-year Treasury has risen more than 110 basis points this year. This is crucial.

Because an AI data center financed when money costs 3% is one thing. Financing hundreds of billions of dollars of AI infrastructure when long-term money costs 5% or more is something very different. And now we arrive at our dot-com comparison.

2026 AI VS. 4Q99-2Q00 DOT-COM

Late 1999 and early 2000 looked fantastic. That’s important. The economy wasn’t falling apart. It was booming at the edges. Real GDP grew roughly 4.2% in 1999. Technology investment was surging. Telecom capital spending exploded.Sound familiar?

THE AI NUMBER IS STAGGERING

Goldman estimates hyperscaler capital spending at roughly $808 BILLION in 2026. That’s up approximately 96% in ONE YEAR. And 2027? About $1.16 TRILLION. Another 44% increase. By 2028, consensus is roughly $1.29 trillion. That’s enormous. But here’s where October gave us an important warning.

The rate of growth is peaking. Goldman says AI-infrastructure earnings growth is currently running around 80%. But consensus sees that slowing toward roughly 40% by late 2027. Again: still enormous growth. But slower.

HERE’S THE DOT-COM PARALLEL THAT BOTHERS ME

We’ve had discussed sin ce late 2qtr 2025 the Dotcom parrels in a positive manner.  We called the late 1q26 decline in stocks the 7^th inning stretch and the subsequent rally likely entering the 9^th inning of the capex AI stock trade. Reca;; back in late 1999, capital spending kept accelerating even as the cost of money increased. The 10-year Treasury rose sharply into early 2000. The Fed was tightening too. Oil surged. Inflation pressures increased. The economy remained strong. Technology capex remained enormous. And stocks initially didn’t care. The Nasdaq gained an astonishing 49% in Q4 1999, then another roughly 9% in Q1 2000. There wasa  cherry on top of tech stock at the end of Dotcom from the Fed increasing liquidity into Y2k which turned into a non event except the liquidity flooded into stocks. And then – the economics of the investment boom began to matter. Q2 2000: Nasdaq about -12%. And the damage was just beginning.

BUT HERE IS THE BIG DIFFERENCE

Today’s AI companies are not Pets.com. Microsoft makes enormous profits. Alphabet makes enormous profits. Meta makes enormous profits. Amazon makes enormous profits. Nvidia makes enormous profits.

Today’s leading AI companies generally have much stronger fundamentals than the speculative leaders of 2000. And today’s S&P 500 trades around 19 times forward earnings. Information Technology also trades around 19 times. That isn’t cheap. But it is nowhere near the valuations of March 2000. So this is not 2000 from a valuation point, But something new has happened.

AI HAS CROSSED A FINANCIAL LINE

Goldman says the five major hyperscalers collectively produced NEGATIVE FREE CASH FLOW in Q3.

About negative $31 billion. Capex has now exceeded their combined operating cash flow.

That’s the number I would put a circle around. Because it changes the AI story. For the first three years, the argument was: These companies can afford it. Now the question becomes: What return are they earning on it? The good news? Cloud revenue growth for Amazon, Microsoft and Alphabet accelerated to 54%.

That’s evidence that AI spending is producing revenue. So this is not fiber-optic cable buried in the ground with no customers. At least not yet.

THIS IS THE LINE BETWEEN BOOM AND BUBBLE

And that brings us to the most important distinction. A boom becomes dangerous when capex growth stays high while revenue growth slows, cash flow falls, and interest rates rise. That is the equation that eventually killed the dot-com capex boom. We aren’t there….. yet. But for the first time in this AI cycle, I think we can see the road that could take us there.

THREE NUMBERS TO WATCH

So here’s my takeaway for retirees and near-retirees.  Don’t try to answer: Is AI a bubble – yes or no?

Instead, watch three numbers. NUMBER ONE: EARNINGS. S&P earnings growth is still around 30%. If that remains strong, the broader bull market has fuel. NUMBER TWO: THE 10-YEAR TREASURY. It seems that the S&P 500 can handle Around 5%, companies can handle it. But if yields keep climbing while earnings slow, valuations become harder to defend.  NUMBER THREE: AI FREE CASH FLOW. This may become the most important number of 2027. AI capex can rise from $800 billion toward $1.2 trillion – if the revenue and cash flow follow it. If they don’t? Then our dot-com comparison moves from interesting to uncomfortable.

October may have given us a preview. AI leadership weakened. But the average stock strengthened. Value beat growth. Energy exploded. Breadth improved. And corporate earnings remained strong.

Bespoke makes an important point: today’s strange breadth readings resemble periods seen after the dot-com peak, but the major indexes have not suffered a major technical breakdown, and today’s leading companies have more defensible valuations. So my conclusion today is: yes we might be in 2000 already for the AI cycle, but the cycle need not end the way Dotcom did.

The AI cycle entered the part of the movie where the cost of capital finally matters. And if you’re retired or close to retirement, that’s exactly why diversification matters. Because October reminded us: You didn’t have to own yesterday’s winners to make money. Sometimes the best sign for a bull market isn’t that the leaders keep leading. It’s that the rest of the market finally starts running with them. And that’s what I’ll be watching next.

Sources: supplied Bespoke, Franklin Templeton and Goldman Sachs reports plus historical comparison research used in the script

 

Do you need a retirement plan that goes beyond allocating funds to truly fit your needs? We can help you create a retirement life plan customized for your retirement vision and legacy. Call us at 877-896-0040 or fill out this form for a free visit: https://click2retire.com/lets-connect