AI Spending Faces Its Verdict: Amazon, Intel and Alphabet Put $220 Billion of Capex on Trial

The AI trade just got its first real cross-examination, and the witnesses did not all help the defence. On Thursday, October 29, US equities ended lower, with the S&P 500 falling 0.30 percent, the Nasdaq declining 1.40 percent and the Dow Jones losing 135 points. The proximate cause was earnings, but the deeper cause was a question investors had been postponing all year: how much cash can the world's biggest companies burn on artificial intelligence infrastructure before the market demands proof that it pays? Alphabet's miss started the interrogation. Amazon and Intel, reporting after the close, extended it. By Friday morning, the verdict was not yet guilty — but the trial had clearly begun.

Take the evidence in order. Alphabet plunged 12.60 percent in Thursday's regular session after its third-quarter earnings per share of $2.12 missed the $3.02 expected, revenue also fell short, and free cash flow was negative for a second straight quarter at minus $9.07 billion. One market review put the single-day market value loss at about $500 billion. Microsoft, by contrast, lost only 0.20 percent despite solid cloud sales — the market distinguishing, at least for a day, between cloud growth that is already monetised and AI spending that is still a promise. Then came the after-hours evidence. Amazon fell 4.20 percent in the regular session and about 6 percent in extended trading after third-quarter earnings and revenue came in below Wall Street estimates, with negative free cash flow of $42.4 billion while maintaining its 2026 capital expenditure forecast of $220 billion. Intel plunged about 21 percent in extended trading — about 17.6 percent by 4:25 p.m. ET — after guiding fourth-quarter revenue to $15–16 billion, below the $17 billion analysts expected, with revenue from its client computing and personal AI group down 5 percent year on year to $8.1 billion. SanDisk slipped 3 percent in extended trading despite beating fiscal first-quarter expectations, a reminder that in this tape, even a beat is not always enough.

The pattern across those reports is what makes this a stocks story rather than a single-stock story. The market is no longer rewarding AI spending as an end in itself; it is asking what the spending buys, and when. Alphabet's negative free cash flow, Amazon's $42.4 billion negative free cash flow against a $220 billion capex plan, and Intel's weaker guidance all point at the same tension: capital expenditure is running far ahead of the cash it generates, and the companies' own forecasts are not closing the gap fast enough for investors who have already priced in years of AI growth. That is why the damage spread beyond the hyperscalers: Marvell dropped 13.24 percent, Micron tumbled 4.98 percent, and Asian technology stocks followed on Friday, with Advantest down 5.43 percent, SoftBank down 4.66 percent and Tokyo Electron down 3.93 percent. When the AI leaders disappoint, the whole supply chain reprices.

It is equally important to note where money went instead, because rotation is the other half of this story. Caterpillar tumbled 9.40 percent after forecasting only moderate fourth-quarter sales growth — a cyclical name punished for a soft outlook — but Eli Lilly rose 6.50 percent after lifting its revenue outlook on strong demand for weight-loss drugs, Merck advanced 3.70 percent after raising guidance, and Shell advanced after lifting its quarterly buyback to $5 billion from $3 billion on stronger refining margins and higher earnings. Healthcare and energy, the least AI-dependent corners of the market, were Thursday's winners. That rotation, plus the month-end scoreboard — the Nasdaq on track to lose 2.5 percent for October, while the S&P 500 is poised to gain 0.9 percent and the Dow to advance 2.8 percent — says the market is not abandoning equities. It is abandoning a single thesis: that AI spending, at any price, justifies any valuation.

The macro backdrop, for once, is not the villain. US third-quarter GDP grew at a 3.4 percent annual rate, beating the 3.2 percent consensus, September headline PCE inflation rose to 3.60 percent year on year from 3.40 percent in August with core PCE at 3.1 percent, the Fed held rates at 3.75 to 4.00 percent on October 28, and jobless claims unexpectedly edged lower to 194,000 in the week through October 24 from a revised 195,000. The 10-year Treasury yield sat at 4.9880 percent. Strong growth, sticky-but-expected inflation and a steady Fed should be a constructive backdrop — and the fact that stocks fell anyway, on earnings rather than on data, is precisely the point. This sell-off was company-specific, and company-specific sell-offs have company-specific remedies: either the next quarter's cash flow justifies the capex, or the capex forecast comes down.

What to watch next, in order. First, Friday's regular session for Amazon and Intel: after-hours moves of 6 percent and 21 percent do not always survive contact with the opening bell, but they set the tone for the whole technology sector into the weekend. Second, Friday's earnings from Exxon Mobil, Chevron, Regeneron and AbbVie — energy and healthcare names that will test whether Thursday's rotation has legs. Third, whether any hyperscaler revises its capex language before year-end; a $220 billion forecast maintained in the face of negative free cash flow is a statement, and the market will treat the next statement as a test of credibility. And fourth, the calendar beyond earnings: US job openings (JOLTS) on November 3. The AI spending trial is not over. But after Thursday, the burden of proof has shifted — from sceptics who had to explain why AI would not pay, to spenders who now have to show, in cash, that it does.

FAQ:
What happened to US stocks on Thursday, October 29? The S&P 500 fell 0.30 percent, the Nasdaq declined 1.40 percent and the Dow lost 135 points, led lower by technology shares after disappointing hyperscaler earnings.
How badly did Alphabet miss? Third-quarter EPS of $2.12 versus $3.02 expected, revenue also short, negative free cash flow of $9.07 billion for a second straight quarter, shares down 12.60 percent, with about $500 billion of market value lost in a day.
What did Amazon report? Third-quarter earnings and revenue below estimates, negative free cash flow of $42.4 billion, while maintaining a 2026 capex forecast of $220 billion; shares fell about 4.20 percent in the session and about 6 percent in extended trading.
What did Intel guide? Fourth-quarter revenue of $15–16 billion, below the $17 billion expected, with client computing and personal AI group revenue down 5 percent year on year to $8.1 billion; shares plunged about 21 percent in extended trading.
Who rose on Thursday? Eli Lilly rose 6.50 percent after lifting its revenue outlook on weight-loss drug demand, Merck advanced 3.70 percent after raising guidance, and Shell advanced after lifting its buyback to $5 billion from $3 billion.
What does the month look like? The Nasdaq is on track to lose 2.5 percent in October, while the S&P 500 is poised to gain 0.9 percent and the Dow to advance 2.8 percent — rotation, not a market-wide collapse.
Is this the same as blog 452? No. Blog 452 covered Alphabet's plunge and the Nasdaq's 1.4 percent fall as a single-stock event. This blog covers the wider AI-capex verdict across Alphabet, Amazon and Intel, plus the rotation into healthcare and energy and the month-end scoreboard.

Related reading:
https://zimbuck.com/pages/zimbucknews
https://zimbuck.com/groups/zimbuckmarketnews
https://zimbuck.com/blogs/452/Alphabet-s-13-Plunge-Dragged-the-Nasdaq-Down-1-4

This is market news and general information only, not financial advice.