The AI Boom Faces a Harder Test

Record results, falling shares and a market asking tougher questions

Nvidia reported another quarter of extraordinary growth. Anthropic's shares surged on their first day of trading. Yet Wall Street ended Thursday with a sixth straight session of losses.

That contrast captures the week's central tension. A business can be thriving while its stock becomes a harder sell.

Yesterday, the SPX500 fell 2.15% to 7,168.36. The NAS100 lost 2.47%, while the US30 dropped 2.30%, shedding more than 1,000 points. The decline was broad, with more than 90% of SPX500 constituents finishing lower.

Nvidia's results made the market's mood especially striking. The chipmaker reported fiscal third-quarter revenue of $115.8 billion, up 103% from a year earlier. Data-centre revenue rose 110% to $107.6 billion, and net income more than doubled to $65.3 billion. Management forecast fourth-quarter revenue of about $130 billion, though that quarter includes an extra week.

Those are formidable numbers. They did not stop Nvidia's shares from falling 3.3% on Thursday, extending their losing streak to six sessions. Investors were also weighing leadership uncertainty following Jensen Huang's resignation, restrictions affecting China and concerns about how some customers are financing their spending.

Trade the News: View our Economic Calendar

Anthropic's IPO offered a similar picture. The AI developer raised about $81 billion after pricing its shares at $83. They opened at $111.50, then fell back to $97.17, still 17% above the offer price. Demand for AI exposure was clearly strong. But the debut did little to lift the wider technology sector.

The market's focus appears to be shifting. Investors are asking whether the enormous spending behind AI will produce returns that justify its cost.

Rates and financing are part of the story

The Federal Reserve's October meeting minutes, released Wednesday, added to the pressure. Policymakers showed growing concern about persistent inflation. Almost all expected further tightening if conditions developed as they anticipated, while some thought a larger increase might eventually be needed. In October, the Fed had held its benchmark rate at 3.75% to 4.00%.

The minutes also raised questions about financial stability. Officials pointed to growing debt tied to AI infrastructure, leasing commitments outside company balance sheets and financing links between firms. Those arrangements could become a weakness if the returns on AI investment fall short.

The concern is straightforward. Data centres require huge sums for processors, electricity, cooling and buildings, often years before the resulting capacity brings in enough cash to cover the cost.

Higher rates make that funding more expensive. They also reduce the value investors assign to earnings expected far in the future, while making government bonds more attractive relative to high-priced growth stocks.

Chicago Fed President Austan Goolsbee added a further complication on Thursday. He warned that the scale of AI investment could strain economic capacity and add to inflation before the technology's productivity benefits become clear.

Oil offered some relief. UKOil traded near $92.50 and was heading for a weekly decline of about 4%, as Middle Eastern shipments recovered and concerns about excess supply weighed on prices. But attacks and disruption risks around the Strait of Hormuz remain a threat to energy flows.

Cheaper oil could eventually help ease inflation. For now, markets seem more focused on the prospect of another Fed increase in December and the financing burden that would bring for companies spending heavily on growth.

The test is whether spending turns into cash

Nvidia's results show that demand for AI computing remains substantial. They don't, by themselves, show that every company buying that computing power will earn an attractive return.

Chipmakers record revenue when they sell equipment. Their customers then have to make that investment pay off over time, while covering depreciation, power costs and financing expenses.

That calculation becomes more fragile when customers depend on borrowing or complex lease arrangements. Construction delays, underused capacity or weaker-than-expected revenue could quickly erode the returns companies have projected.

Over the next 6 to 24 months, investors will want to see whether capital spending is matched by operating cash flow. Free cash flow, borrowing costs and returns on invested capital will matter too. So will evidence that AI products can generate lasting commercial revenue, not just demand for more computing capacity.

There is a credible case for optimism. Businesses may adopt AI faster than expected, efficiency may improve and productivity gains may eventually justify today's spending. Nvidia's results suggest that demand for the underlying infrastructure has not vanished.

But the market is beginning to draw a sharper line between technological promise and financial proof. This week's sell-off does not establish that AI is a bubble. It does suggest that investors are becoming less willing to pay premium prices without clearer evidence of sustainable returns.

AI may keep advancing at remarkable speed. The harder question is whether company earnings and cash flows can grow fast enough to justify the money already committed. That is becoming the defining investment test for the boom.

References and sources

Russell Shor

Senior Market Strategist

Russell Shor is a Senior Market Strategist at FXCM, having been promoted to the role in 2025 in recognition of his depth of insight and consistent delivery of high-impact market analysis. He originally joined FXCM in October 2017 as a Senior Market Specialist.

Russell holds an Honours Degree in Economics from the University of South Africa, is a certified FMVA®, and a full member of the Society of Technical Analysts (UK). With over 20 years of experience in financial markets, his work is renowned for its clarity, precision, and strategic value across asset classes.

${getInstrumentData.name} / ${getInstrumentData.ticker} /

Exchange: ${getInstrumentData.exchange}

${getInstrumentData.bid} ${getInstrumentData.divCcy} ${getInstrumentData.priceChange} (${getInstrumentData.percentChange}%) ${getInstrumentData.priceChange} (${getInstrumentData.percentChange}%)

${getInstrumentData.oneYearLow} 52/wk Range ${getInstrumentData.oneYearHigh}
Disclosure

Any opinions, news, research, analyses, prices, other information, or links to third-party sites contained on this website are provided on an "as-is" basis, as general market commentary and do not constitute investment advice. The market commentary has not been prepared in accordance with legal requirements designed to promote the independence of investment research, and it is therefore not subject to any prohibition on dealing ahead of dissemination. Although this commentary is not produced by an independent source, FXCM takes all sufficient steps to eliminate or prevent any conflicts of interests arising out of the production and dissemination of this communication. The employees of FXCM commit to acting in the clients' best interests and represent their views without misleading, deceiving, or otherwise impairing the clients' ability to make informed investment decisions. For more information about the FXCM's internal organizational and administrative arrangements for the prevention of conflicts, please refer to the Firms' Managing Conflicts Policy. Please ensure that you read and understand our Full Disclaimer and Liability provision concerning the foregoing Information, which can be accessed here.

Past Performance: Past Performance is not an indicator of future results.