2026-10-10

Dow Jumps 423 Points, S&P 500 and Nasdaq Rally Friday as Wall Street Shakes Off Its Worst AI Selloff Since August

Written & reviewed by the TradeSmrt editorial team · 2026-10-10

In this article

What Happened
Why a Single Session Doesn't Settle the AI Valuation Debate
What to Take Away From This
FAQ
Sources

What Happened

Wall Street closed out a choppy week with its strongest session in days on Friday, October 9. The Dow Jones Industrial Average jumped 423.31 points, or 0.83%, to finish at 51,654.95. The S&P 500 gained 0.59% to close at 7,811.54, and the Nasdaq Composite rose 0.64% to 27,366.17. All three major indexes clawed back most of what they had lost just one session earlier, when the Nasdaq fell 1.3% on Thursday - its sharpest single-day drop since mid-August - as AI-linked chip and software stocks led a broad pullback.

Thursday's slide traced back to a narrow, specific catalyst rather than any broad economic warning sign: a Financial Times report said OpenAI's annualized revenue stood at roughly $50 billion as of the end of September, nearly $20 billion below the "nearing $70 billion" figure that had circulated widely across markets just weeks earlier, after a Reuters report in late September cited similar growth. Oracle (NYSE: ORCL), whose stock has become closely tied to its multi-year cloud-computing contracts with OpenAI, fell 5.48% on the news. Intel (NASDAQ: INTC) dropped nearly 6%, Broadcom (NASDAQ: AVGO) slid 4.35%, Nvidia (NASDAQ: NVDA) lost 2.94%, and AMD fell more than 4% as well. The damage was concentrated enough in AI infrastructure names that the Dow, which carries less semiconductor weight, actually closed Thursday essentially flat while the Nasdaq absorbed the brunt of the selling.

By Friday, that same cohort of AI-exposed stocks found a floor and helped pull the broader market higher. Trading desks and market commentary pointed to a simple realization sinking in overnight: the "$20 billion gap" reflected a difference in accounting convention rather than an actual slowdown in demand. OpenAI's reported figures count only its own net revenue, while the widely cited $70 billion estimate had effectively been built by applying rival Anthropic's growth-rate math - a company that, unlike OpenAI, includes revenue booked through cloud partners such as AWS and Google Cloud at those partners' own billing rates. Once that distinction was better understood, and with OpenAI's enterprise segment still reportedly growing well over 100% year over year by the Financial Times' own figures, some of Thursday's reflexive selling looked overdone by Friday morning. Risk appetite recovered elsewhere too: bitcoin traded near $83,000, up roughly 1.7% on the day, tracking the broader shift back toward risk assets.

Importantly, Friday's rebound happened against an unchanged - not improved - backdrop in the bond market. Just two sessions earlier, on October 7, the 10-year Treasury yield had touched 5.365% intraday, its highest level since April 2002, while the 30-year yield climbed as high as 5.732%, a level not seen since 2002. Those yields, driven by a mix of oil-linked inflation worries, heavy federal bond issuance, and growing unease about the scale of AI-related corporate debt, stayed near those multi-decade highs through the rest of the week. In other words, stocks rallied Friday not because financing conditions eased, but because one specific point of investor confusion got cleared up. Strategists quoted in Friday's market coverage cautioned that AI-linked names were unlikely to stage a sharp "V-shaped" recovery given how much uncertainty remains around financing costs and revenue verification across the sector. Still, despite Thursday's dip, all three major averages ended the week higher overall - a reminder of how quickly sentiment has been swinging from session to session this month.

Why a Single Session Doesn't Settle the AI Valuation Debate

It's tempting to read Friday's bounce as proof that Thursday's selloff was simply an overreaction, and in a narrow sense it was - the revenue dispute turned out to be a matter of accounting definitions, not an actual business slowdown. But the deeper question that investors were really pricing earlier in the week - whether the AI infrastructure buildout can comfortably support the enormous volume of corporate debt being issued to fund it - was not resolved by one good trading day. Oracle's credit default swaps, a market-based gauge of how risky bond investors judge the company's debt to be, remained near their highest levels since 2009 even as the stock price itself recovered. Meta, Oracle and other major AI spenders have together raised well over $100 billion in bond markets in 2026 specifically to fund data centers and chip purchases, and persistently high long-term Treasury yields raise the cost of refinancing that debt for years to come, regardless of what any single week's revenue headline says.

This gap between what moved the stock price and what actually changed is a useful distinction for anyone following rate-sensitive, high-growth sectors. A stock's near-term price action is often driven by headline risk and short-term positioning - how quickly investors reacted to one data point, and how quickly they reconsidered it. Its underlying valuation risk, by contrast, is driven by a slower-moving set of structural factors: debt loads, financing costs, and the verifiability of the revenue projections that justify the spending in the first place. When a sharp move happens on a single data point, in either direction, it's worth asking which of these two things actually happened. Here, a specific piece of confusion got cleared up, which was genuinely good news - but the larger question of whether this capital spending cycle can be sustainably financed at current yield levels remains exactly as open as it was on Wednesday.

What to Take Away From This

Don't confuse a sentiment-driven bounce with a resolved risk. Friday's rally cleared up one specific point of confusion about accounting methodology, but it didn't change the structural backdrop - near-24-year-high bond yields and heavy AI-related debt issuance - that triggered the broader selloff just two days earlier.
Watch credit markets, not only stock prices, for the fuller picture. Oracle's CDS spreads staying elevated even as its share price recovered is a sign that bond investors are pricing the underlying risk differently than equity traders reacting to a single news cycle.
Volatility in a concentrated, high-momentum sector can run in both directions within the same week. The same AI-linked names that led Thursday's decline led Friday's recovery, which is a useful reminder that single-session moves in crowded trades often overshoot on the way down and on the way back up.
A strong headline index number doesn't mean every stock inside it moved equally. As with earlier in the week, Friday's gains were concentrated in a relatively narrow group of large-cap tech and chip names, even though the index-level numbers looked uniformly positive.

FAQ

Does Friday's rally mean concerns about an AI spending bubble are over?

No. The rally mainly resolved a specific accounting-methodology dispute over how OpenAI's revenue should be counted, not the broader question of whether AI-related companies can sustainably service the debt they've taken on to fund data centers and chips. Credit-market indicators like Oracle's credit default swaps stayed elevated even as the stock price bounced back, which suggests bond investors still see unresolved risk that a one-day equity rally didn't erase.

Why did stocks rise even though Treasury yields stayed near 24-year highs?

Friday's move and the bond market move were responding to different things. The elevated yields reflect longer-running concerns about inflation, heavy government debt issuance, and oil prices that didn't change much in a single day. Friday's equity rally was driven by a specific, resolvable piece of news - the OpenAI revenue discrepancy - showing that a stock rebound and an unresolved macro backdrop can coexist in the same session.

Why did the Dow underperform the Nasdaq on Thursday but outperform it on Friday?

The Dow has much lighter exposure to semiconductor and AI-infrastructure stocks than the Nasdaq does, so it barely moved on Thursday's selloff, which was concentrated in chip and cloud names. On Friday, several Dow components outside the AI trade also advanced, which is why its percentage gain (0.83%) ended up larger than the Nasdaq's (0.64%) even though the Nasdaq had fallen further the day before.

Sources

This article is an original synthesis and analysis based on the following reporting; it does not reproduce any source's text.

This article is for informational and educational purposes only and does not constitute investment advice. All investments carry risk, including the potential loss of principal. Past performance and historical patterns do not guarantee future results. Always do your own research and consult a licensed financial advisor before making investment decisions.