The real question is how long the spending will last and how bad the fallout will be when it ends.
In the middle of November 2025, Asian markets paused in uncertain equilibrium as a three-day global selloff forced investors to confront a question that had long gone unasked: whether the extraordinary faith placed in artificial intelligence as an economic engine had outrun the evidence. The S&P 500's worst monthly performance since 2008 and Nvidia's stumble ahead of earnings revealed not merely a correction in prices, but a reckoning with the distance between technological promise and measurable return. Markets, like civilizations, must eventually reconcile their visions with their ledgers — and that moment of reconciliation appeared to have arrived.
- A three-day global selloff has left the MSCI Asia Pacific at its worst stretch since April, with the S&P 500 down over 3% in November — its bleakest month since the 2008 financial crisis.
- The AI rally's foundational assumption — that massive infrastructure spending would eventually justify itself — is now openly questioned, with Wall Street's fear gauge climbing above 24 and Nvidia falling 2.8% ahead of a closely watched earnings report.
- Federal Reserve rate-cut expectations have quietly collapsed, with swap markets now pricing in less than a 50% chance of a December cut, as some officials warn of lingering inflation risks.
- A delayed US jobs report, due Thursday, has become a pivot point: a weak reading could revive rate-cut hopes and ignite a year-end rally toward S&P 7,100, while a strong one may deepen the uncertainty.
- Nvidia's earnings — for a company now larger than the energy, materials, and real-estate sectors combined — have become a referendum on whether the entire AI investment thesis can survive contact with reality.
Asian markets opened Wednesday in a fragile holding pattern, absorbing the aftershocks of a three-day global selloff that had left the MSCI Asia Pacific at its worst stretch since April. The damage had originated in New York, where the S&P 500 had fallen for four consecutive sessions and the so-called Magnificent Seven mega-cap tech stocks had dropped 1.8% as a group. Nvidia, the chip maker most closely identified with the AI boom, slid 2.8% ahead of its earnings report, while US futures edged lower in early Asian trading.
What had unsettled investors was a deeper, more philosophical doubt: after months of accepting enormous AI infrastructure spending as the cost of building the future, markets were now demanding to see the returns. The Cboe Volatility Index climbed above 24, the S&P 500 was on pace for its worst November since 2008, and even Bitcoin had wobbled below $90,000 before steadying. Analyst Sonu Varghese framed the central anxiety plainly — not whether a bubble existed, but how long companies would keep spending before demanding profits, and how violently markets would react when that spending slowed.
A parallel unease surrounded the Federal Reserve. Swap markets had shifted to imply less than a 50% probability of a December rate cut, reversing earlier expectations, as some officials cautioned against moving too quickly. Treasury yields slid as investors sought safety, and Treasuries were on track for back-to-back monthly gains. The September jobs report, delayed by a historic government shutdown, was due Thursday — and a weaker-than-expected reading, analysts suggested, could reopen the door to a December cut and a year-end rally pushing the S&P 500 toward 7,100.
Nvidia's singular weight in the market had grown almost difficult to comprehend — larger than the energy, materials, and real-estate sectors combined, and bigger than all of industrials. Its earnings would send ripples through markets worldwide. Elsewhere, Microsoft and Nvidia announced a combined $15 billion investment in AI developer Anthropic, oil rose on tightened Russia sanctions rhetoric, and Asian corporate results told an uneven story: Baidu posted its worst quarterly revenue decline on record, PDD Holdings warned of slowing growth in China's e-commerce wars, but Xiaomi reported its first quarterly profit from electric vehicles — a small, quiet milestone amid the noise.
Asian markets opened Wednesday in a holding pattern, caught between small gains and modest losses as traders tried to read the damage from three days of selling that had rippled across the globe. The MSCI Asia Pacific index slipped 0.1%, a relatively gentle move that masked the real story: the region's largest benchmark had just posted its worst stretch since early April, and no one was quite sure where the bottom was.
The trouble had started in New York. The S&P 500 had now fallen for four consecutive days, with the so-called Magnificent Seven—the handful of mega-cap tech stocks that have carried the market's weight all year—dropping 1.8% as a group. Nvidia, the chip maker that has become synonymous with the artificial intelligence boom, fell 2.8% in anticipation of earnings due after Wednesday's close. Futures on US stocks edged lower in early Asian trading, suggesting the selling might not be finished.
What had spooked investors was a creeping doubt about the entire premise of the AI rally. For months, the market had accepted massive spending on AI infrastructure as an article of faith—the cost of building the future. But now, after the longest stretch of gains, traders were asking a harder question: where were the profits? Wall Street's fear gauge, the Cboe Volatility Index, had climbed above 24, well past the 20 level that signals real anxiety. The S&P 500 was down more than 3% for November, on pace for its worst month since 2008. Bitcoin, which had briefly dipped below $90,000 on Tuesday, had steadied around $92,500, but the wobble had been enough to unsettle sentiment across risk assets.
Sonu Varghese, an analyst at Carson Group, captured the shift in thinking. The question wasn't whether a bubble existed, he said. The real question was how long companies would keep spending on AI infrastructure before demanding returns, and how badly markets would react when that spending finally slowed. The answer to that question would likely come from Nvidia's earnings report—a single company's results that had somehow become a referendum on the entire technology sector and, by extension, the broader market's ability to sustain its gains.
Meanwhile, a separate anxiety was building around the Federal Reserve. Traders had grown less confident that the central bank would cut interest rates again in December. Swap markets now implied less than a 50% probability of a rate cut next month, a sharp reversal from earlier expectations. Some Fed officials had recently warned against another cut, citing inflation risks, though Governor Christopher Waller had continued to argue for lower rates. The yield on 10-year Treasury bonds had slid three basis points to 4.11% as investors rotated into safer assets, and Treasuries were on track for their first back-to-back monthly gains.
The labor market, too, was sending mixed signals. Jobless claims had totaled 232,000 in the week ended October 18, while companies had shed an average of 2,500 jobs per week in the four weeks ended November 1, according to ADP Research. The September jobs report was scheduled for release Thursday, delayed by the longest government shutdown in history. If that report came in weaker than expected, James Demmert at Main Street Research suggested it could clear the way for a December rate cut and fuel what he called a "Santa Claus rally"—a year-end surge that could push the S&P 500 to 7,100.
The scale of Nvidia's importance to markets had become almost absurd. The company was now larger than the energy, materials, and real-estate sectors combined, and on some days exceeded the combined weight of those three plus utilities. It was bigger than the entire industrials sector. Ryan Grabinski at Strategas noted that Nvidia's earnings would likely send ripples through both US and international markets. Expectations for AI had cooled in recent weeks, he said, but a strong earnings report could shift sentiment back toward optimism—though the bar for that optimism was undeniably high.
In other corners of the market, Microsoft and Nvidia announced they would invest up to $15 billion combined in Anthropic, an AI developer, tying the company closer to two of the biggest backers of OpenAI. Oil rose on hawkish rhetoric from the European Union about tightening sanctions on Russia. Gold wavered. And in Asia, corporate earnings painted a picture of uneven momentum: Baidu posted its biggest quarterly revenue decline on record as its advertising business weakened; PDD Holdings warned of a slowdown in China's intensely competitive e-commerce market; but Xiaomi reported its first quarterly profit from its electric vehicle business, marking a milestone in its ambitious push into that crowded sector.
Notable Quotes
The question isn't really whether we're in a bubble. The real question is how long the current trend in AI spending will last and how bad the fallout will be when it ends.— Sonu Varghese, Carson Group
The outcome is likely to send ripple effects through both US and international markets. Although expectations for AI more broadly have cooled in recent weeks, this report has the potential to shift sentiment back to optimism. That said, the bar is undeniably high right now.— Ryan Grabinski, Strategas
The Hearth Conversation Another angle on the story
Why does one company's earnings report matter so much to the entire market?
Because Nvidia has become the physical embodiment of the AI bet. It makes the chips everyone needs to build AI systems. If Nvidia's earnings disappoint, it suggests the whole infrastructure spending spree isn't generating returns yet—and that's the foundation the entire rally has been built on.
So what happens if the earnings are weak?
The selling probably continues. But if they're strong, it could convince traders that the AI spending is actually working, and you might see a sharp reversal—what some analysts are calling a year-end rally that could push the market significantly higher.
What about the Fed? Why does that matter right now?
Because if the Fed cuts rates in December, it makes borrowing cheaper and typically helps stocks. But traders aren't sure the Fed will do it. If the jobs report Thursday shows weakness, it might force the Fed's hand and give markets the rate cut they've been hoping for.
Is this a bubble?
That's what everyone's asking. The real question isn't whether it's a bubble—it's whether companies can actually make money from all this AI spending before the market demands returns. Right now, that's still uncertain.
What's the risk if nothing changes?
The selling continues. Markets have already had their worst month since 2008. If Nvidia disappoints and the jobs report stays weak, you could see more downside before any recovery takes hold.