U.S.-China Relations
NextFin News - A single trading session on October 29 laid bare the three uncertainties now defining the end of 2026: Intel's stock collapsed more than 20% after hours despite beating third-quarter revenue and earnings estimates, Amazon's record cloud growth was drowned out by a capital-spending plan that has pushed free cash flow deeply negative, and the Justice Department escalated its fight over California's election administration by issuing grand jury subpoenas to local officials five days before Election Day. The common thread is a market and a political system shifting from growth narratives to accountability — for AI dollars, for guidance credibility, and for the integrity of the vote itself.
The Day in Three Numbers
Intel reported third-quarter revenue of $16.5 billion, up 21% from a year earlier, with non-GAAP earnings per share of $0.52 — ahead of the consensus EPS estimate of roughly $0.40 and the company's own July guidance. The stock still finished the regular session down 3.6% at $92.07, then plunged to $70.71 in after-hours trading, a 23.2% wipeout. The trigger was not the quarter that passed but the one ahead: Intel forecast fourth-quarter revenue of $15.0 billion to $16.0 billion and non-GAAP EPS of $0.25, below the midpoint of its prior $15.8 billion to $16.8 billion range and the $0.38 per-share guide issued in July.
Amazon, reporting after the close, posted net sales of $201.9 billion, up 12% year over year and just under the $202.2 billion consensus, with operating income of $28.1 billion — a record 13.9% operating margin. AWS revenue accelerated 39.3% to about $46 billion, its fastest growth in nearly five years, taking the cloud unit to a $184 billion annualized run rate. Diluted EPS came in at $1.89 versus expectations near $1.96. Shares fell 4.3% during the regular session and gave up another 6.6% after hours, as investors focused on management's confirmation that 2026 capital expenditure will run about $220 billion — up from the $200 billion guided in July — and on a fourth-quarter sales forecast of $240 billion to $247 billion that trailed the $244.5 billion consensus.
Separately, the Justice Department issued federal grand jury subpoenas to state and local election officials in California as part of a widening criminal investigation into allegations that as many as 20,000 non-citizens are registered to vote in the state. The subpoenas, sent between October 21 and October 24, came one day after U.S. District Judge David Carter rejected the administration's bid to force California officials to set aside mail ballots received from roughly 20,000 people it claims are not U.S. citizens, calling the move a violation of "longstanding constitutional principles" giving states authority over their own elections. More than 3.5 million mail ballots have already been returned statewide ahead of the November 3 vote.
Intel: A Beat That Priced the Wrong Quarter
The market's verdict on Intel was brutal precisely because the quarter itself was not weak. The Data Center and AI group, the company's most important growth engine, saw revenue jump 78% to $7.3 billion — the strongest growth in 15 years — while Intel Foundry revenue rose 58% to $6.7 billion. The company also completed a $23 billion public equity offering, strengthening a balance sheet it is leaning on to fund a foundry buildout that remains years from profitability. Chief executive Lip-Bu Tan called the results evidence of "improving execution, meaningful progress expanding supply and continued advancement of our process technology."
But the guidance told a different story. CFO David Zinsner said the company expects the PC slowdown to continue "at least through the first half of next year." Client Computing revenue fell 5% to $8.1 billion in the quarter, and the fourth-quarter revenue guide implies a sequential contraction from the $16.5 billion just reported. For a stock that has rallied on the promise of an AI-driven recovery, the message was that the recovery is lopsided: data centers are booming, PCs are softening, and the foundry turnaround is still a cash consumer rather than a cash generator.
"We delivered another solid quarter. Revenue, gross margin and earnings per share exceeded guidance, including our strongest data center revenue growth in 15 years," said David Zinsner, Intel's chief financial officer. "Looking ahead, we see the PC slowdown continuing at least through the first half of next year."
The options market had priced a ±10.7% move for the report as of October 23; the actual after-hours swing blew past it. That gap matters. When a company beats the current quarter and still gets punished, the market is telling you it no longer trusts the forward narrative enough to look through a weak guide. Intel's challenge is now one of credibility as much as execution: every future quarter will be read against a management team that has reset expectations before.
The cyclical-versus-structural call: the PC downturn is cyclical — inventories clear, refresh cycles return, and mean reversion is the base case. The foundry bet is structural, and it is not yet proven. Those two forces point in opposite directions, which is exactly why the stock is uninvestable for many until the foundry path shows a credible route to positive cash flow.
Amazon: The AI Arms Race Becomes a Cash-Flow Story
Amazon's quarter was, on its face, excellent. Operating margin hit a record, AWS growth re-accelerated to levels not seen since 2021, and CEO Andy Jassy projected AWS revenue growth above 40% in both 2027 and 2028.
"Our AWS investment in functionality and infrastructure is working, with over 39% year-over-year revenue growth in Q3 ($184 billion revenue run rate) and projected revenue growth rates of over 40% both in 2027 and in 2028," said Andy Jassy, Amazon's chief executive officer.
He has also said AWS capacity is largely sold out for next year, with a good chunk of 2028 already committed — a demand backdrop most companies would kill for.
Yet the market sold the stock. The reason is the denominator of the AI equation, and the cash-flow statement makes it stark. For the trailing twelve months through September 30, Amazon generated $168.3 billion of operating cash flow but spent $215.2 billion on property and equipment, producing a free-cash-flow outflow of $42.4 billion — a swing from a $14.8 billion inflow a year earlier. In the first nine months of 2026 alone, capital spending reached $175.7 billion, nearly double the $92.3 billion in the same period a year earlier. Management has confirmed 2026 capital expenditure of about $220 billion, up from $200 billion in July, with analysts surveyed by StreetAccount expecting it to climb past $280 billion in 2027. Amazon has joined the group of hyperscalers for whom AI is no longer a revenue story but a financing story.
This is the second-order shift the market is grappling with. For two years, AI capital expenditure was rewarded on the theory that more compute would compound into more cloud revenue. That theory is now being stress-tested: revenue is indeed compounding — AWS is on a $184 billion run rate — but the cash required to build the capacity is compounding faster. The question investors are now asking is not "how fast is AWS growing?" but "what return does each dollar of capex earn, and when does free cash flow turn back positive?"
The closest analog is Amazon's fulfillment buildout during the pandemic, when capital intensity surged and then normalized as the network matured. The difference this time is scale and duration: a data center is a longer-lived, more specialized asset than a warehouse, and the competitive pressure to keep spending — from Microsoft, Alphabet, and Meta — means Amazon cannot easily pause without risking the very capacity commitments Jassy is touting. That is the trap: stopping spending jeopardizes the growth; continuing it depresses cash flow. The market has not decided which side of that trade Amazon is on.
Cyclical or structural? The capex surge is structural in the sense that the AI infrastructure buildout is a multi-year regime shift, not a one-year spike. But the valuation multiple compression we are seeing is the market repricing that shift from "growth at any cost" to "growth with a return hurdle." That repricing can be cyclical — it can overshoot and recover — if AWS monetization accelerates faster than spending.
California: Election Scrutiny Moves From Civil Court to Criminal Probe
The third thread of the day had nothing to do with earnings and everything to do with institutional trust. The Justice Department's escalation to a criminal investigation — marked by grand jury subpoenas to California election officials — raises the stakes in a legal battle that was already one of the most consequential election fights of the cycle. A federal judge had just blocked the administration's attempt to separate out roughly 20,000 ballots; the response was to open a criminal probe and subpoena the officials who refused.
California officials have pushed back hard. Los Angeles County Registrar-Recorder Dean Logan called the Justice Department's approach "confrontational" and "adversarial," citing the judge's admonishment that "these threats must stop." A Justice Department spokesperson said unlawful voting "is properly addressed through criminal law." The factual dispute underneath is unresolved: a Los Angeles County sample of 200 voters found that just over 38% reported being born in the United States, while a U.S. Citizenship and Immigration Services official said its own review of the same 200 records determined none were U.S.-born, acknowledging separately that about 189 records carried "minor data discrepancies."
With more than 23 million registered voters in California and Election Day five days away, the practical stakes are clear: any move to discount ballots after the fact would be challenged in court, and the window for resolution before certification is narrow. The administration has pursued parallel efforts in Washington state, where prosecutors asked officials to set aside ballots from more than 400 alleged non-citizens, and in Alaska, where it has sued to block certification of a Senate race if the incumbent loses. The pattern is a sustained push to assert federal oversight over election administration in Democratic-led states.
The structural read: this is not a one-off legal skirmish. It is a stress test of the post-2020 election-administration framework — who certifies votes, who can challenge them, and how close to Election Day disputes can be raised. Markets hate uncertainty, and while election outcomes are political rather than portfolio decisions, prolonged disputes over certification can feed volatility into rate and equity markets if the result in a closely divided Congress hangs in the balance.
What Comes Next
Three signals will determine whether today's moves were an overreaction or the start of a repricing.
For Intel, watch the PC refresh cycle and foundry cash burn over the next two quarters. If client revenue stabilizes by mid-2027 and foundry losses narrow toward the company's roadmap, the 23% after-hours drop will look like a buying opportunity. If the PC slowdown extends beyond the first half of next year as management expects, the multiple has further to fall.
For Amazon, the falsifying signal is free cash flow. If AWS revenue growth holds above 40% while capital expenditure growth decelerates in 2027 — a plausible outcome if memory costs normalize and capacity additions moderate — cash flow turns positive and the selloff reverses. If capex keeps climbing past $280 billion while growth decelerates, the market will keep discounting the stock.
For California, watch whether any court rules on the DOJ's subpoena enforcement or ballot challenges before November 3, and whether the administration escalates to threats against certification itself. A quiet election with no successful ballot challenges would defuse the episode; a contested certification would extend the uncertainty well into December.
The through-line of October 29 is accountability. Intel was punished for a guide that failed to match its beat. Amazon was punished for a spending plan that outruns its cash generation. And the Justice Department's California push is a reminder that in an election year, even the institutions that certify outcomes are subject to the same scrutiny as the companies that report them. The market is no longer paying for promises — it is paying for proof.
Explore more exclusive insights at nextfin.ai.
