


Wall Street staged a sharp reversal on Thursday as a blowout earnings report from Microsoft ignited a broad Nasdaq rebound, just one day after the index had extended a punishing selloff tied to fears over the sustainability of artificial intelligence spending. The Nasdaq 100 climbed 3.4% and the S&P 500 rose 1.7%, snapping a six-day losing streak, as investors rotated back into megacap technology names on evidence that at least one hyperscaler's AI investments are translating into accelerating, not decelerating, growth.
Microsoft shares surged as much as 16% in early trading, adding roughly $450 billion in market value in a single session, the largest one-day gain by any company in history in absolute dollar terms. The move came after the company reported fiscal fourth-quarter revenue of $90 billion, up 18% year-over-year and well ahead of the roughly $87.6 billion Wall Street had expected. Adjusted earnings per share came in at $4.74, excluding a one-time gain tied to Microsoft's stake in OpenAI, comfortably ahead of the $4.24 consensus estimate.
The headline figure driving the rally was Azure. Microsoft's cloud computing division grew 43% year-over-year, accelerating from 40% growth the prior quarter and blowing past management's own guidance of 39% to 40%. That acceleration mattered enormously to a market that had spent the previous week fixated on whether AI infrastructure spending was still generating commensurate returns. Azure's annual revenue run rate crossed $100 billion for the first time, a milestone that chief executive Satya Nadella highlighted on the earnings call alongside disclosure that the company had added a full gigawatt of new data center capacity in the quarter alone.
The company is delivering better Azure and finally better Office growth, while also not surprising negatively on its capex outlook and free cash flow targets.
Just as important to investors as the growth acceleration was what Microsoft did not do: it did not raise its already enormous capital expenditure guidance. Quarterly capex rose 70% year-over-year to $41 billion, but that figure came in slightly below the roughly $42.4 billion analysts had expected, and management reaffirmed its full-year framework rather than layering on additional spending commitments. Roughly two-thirds of that capex is going toward short-lived assets such as GPUs and CPUs rather than long-lived construction, a mix investors have come to view as a signal of financial discipline.
The contrast with a rival hyperscaler could not have been sharper. Alphabet's Google had raised its own capital expenditure outlook to $205 billion just days earlier, a disclosure that sent its shares down roughly 6% on concerns about spiraling AI infrastructure costs without commensurate revenue proof points. Microsoft's decision to hold the line on spending while simultaneously delivering accelerating cloud growth was, in the words of one Wall Street analyst team, the clearest reward of the entire earnings season for restraint paired with results.
Microsoft's report acted as the catalyst for a far broader risk-on move across technology stocks. The Nasdaq Composite finished 2.8% higher, ending its own six-day losing streak, while the Dow Jones Industrial Average surged more than 600 points. Semiconductor stocks, which had been at the center of the prior week's selloff, rebounded sharply as well, with a closely watched gauge of chipmakers posting its best day since April 2025 as traders speculated that the broader AI trade still had room to run despite the prior week's volatility.
Amazon added further fuel to the rally in after-hours trading, jumping after reporting that cloud-computing revenue growth at Amazon Web Services had accelerated for a fifth consecutive quarter. That result reinforced the emerging narrative among investors: the two largest US cloud providers were both showing durable, accelerating demand for AI-linked infrastructure, even as skepticism about the broader AI capital-expenditure cycle had driven a sharp Nasdaq rebound-preceding selloff just days earlier.
The rally was not universal. Apple shares slipped in after-hours trading following the release of its own results, as investors focused on weaker-than-expected sales in China, a market that has become an increasingly important swing factor for the iPhone maker's overall growth trajectory. The divergence underscored that this week's Nasdaq rebound was driven less by blanket optimism about technology stocks broadly and more by company-specific evidence that AI-related capital spending is, in at least some cases, converting into durable top-line growth.
Microsoft: shares surged as much as 16% on accelerating Azure growth and disciplined capex guidance
Amazon: shares jumped in late trading after AWS growth accelerated for a fifth straight quarter
Apple: shares slipped after hours on weaker-than-expected China sales
Alphabet: shares had fallen roughly 6% days earlier after raising its capex outlook to $205 billion
The whipsaw price action across AI-linked stocks over the past two weeks has not been costless for everyone. Situational Awareness, a hedge fund launched by former OpenAI researcher Leopold Aschenbrenner with roughly $24 billion in assets, has reportedly been rapidly cutting positions after suffering steep losses tied to both long positions in AI infrastructure names, including South Korean memory maker SK Hynix, and a bearish bet against software stocks such as Adobe that instead rallied. The fund has reportedly been working with prime brokers including Bank of America, Goldman Sachs, and JPMorgan Chase to unwind exposure and meet margin requirements in an orderly fashion.
The episode is a reminder that even as the broader market staged a dramatic Nasdaq rebound, the underlying volatility in AI-linked names over the preceding two weeks was severe enough to force significant deleveraging among funds that had taken concentrated, highly directional bets on how the AI trade would unfold. Prime brokers were reportedly marketing baskets of the fund's long and short positions to potential buyers as part of the unwind, a process that can itself add to volatility in the affected names.
The scale of the reversal is easier to appreciate against the backdrop of what preceded it. Just two trading sessions earlier, the Dow Jones Industrial Average had tumbled more than 1,150 points, its worst single day since April 2025, after the Federal Reserve held interest rates steady and semiconductor stocks extended a decline that had by then stretched into a fifth consecutive losing session. The Philadelphia Semiconductor Index had fallen more than 5% that day alone, with Micron Technology sliding nearly 10% amid fears that Chinese competition and softening AI-linked demand might be catching up with the sector's stretched valuations.
That such a severe, multi-day selloff could be substantially reversed within 48 hours speaks to how binary sentiment around the AI trade has become. Investors were not, in aggregate, revising down their long-term expectations for artificial intelligence adoption during the correction; rather, they were repricing the near-term risk that infrastructure spending might be running ahead of demonstrable returns. Microsoft's results directly addressed that specific concern, which helps explain why the market's reaction was so outsized relative to a single company's earnings report.
Wall Street research desks moved quickly to revise their views following the results. Barclays analysts, who had maintained a more cautious stance on Microsoft through much of the year amid the stock's roughly 19% year-to-date decline heading into the print, said the results gave investors reason to revisit the shares given improving Azure and Office growth alongside a capex outlook that avoided negative surprises. Morgan Stanley, which had maintained an Overweight rating with a $600 price target heading into the report, said the results were consistent with its thesis that Azure growth would accelerate through the back half of calendar 2026 and into fiscal 2027.
The report also carried implications well beyond Microsoft's own stock price. Nvidia, whose GPUs underpin much of the capacity expansion described on Microsoft's earnings call, was seen as a direct beneficiary of the company's reaffirmed capital spending plans, since continued GPU investment from major cloud providers directly supports chip demand. That read-through helped extend the rebound across the broader semiconductor complex, reversing a substantial portion of the prior week's chip-stock losses in a single session.
With Microsoft and Amazon both delivering evidence of accelerating cloud demand, and Alphabet and Meta having already reported in the same stretch, the market now has a nearly complete picture of how the four largest hyperscalers are framing their AI investment cycles heading into the back half of the year. Investors are likely to spend the coming weeks parsing several key questions.
Will Microsoft's Azure acceleration prove durable into fiscal 2027, where management has already guided to roughly 45% constant-currency growth?
Can capital expenditure across the hyperscaler cohort stabilize near current levels without derailing the buildout pace that AI demand appears to require?
Will semiconductor and memory names, which bore the brunt of the prior week's selloff, sustain their rebound as physical component pricing continues to climb?
How much further deleveraging remains among funds that built concentrated positions around the AI trade, and could further unwinds reintroduce volatility?
For now, the Nasdaq rebound has offered at least a partial answer to the doubts that fueled the prior week's correction: at the very least, the largest cloud providers are not seeing AI-related demand slow, even if the price of meeting that demand continues to climb. Whether that reassurance proves durable, or merely a one-earnings-report reprieve, will likely depend on how the remaining details of this earnings season, and the broader trajectory of AI infrastructure spending, continue to unfold.
Portfolio managers who track the technology sector closely noted that the speed and magnitude of this week's reversal, from a sharp correction to one of the strongest single-day rallies of the year, illustrates just how sensitive AI-linked valuations have become to individual earnings prints. A single data point, in this case Microsoft's Azure acceleration, was enough to swing sentiment across the entire sector, a dynamic that cuts both ways: it can drive sharp rebounds just as easily as it can trigger sharp selloffs.
Financial advisors generally caution against extrapolating too much from any single week of price action, however dramatic, and note that the underlying question, whether AI infrastructure spending will ultimately generate a commensurate return on invested capital, remains unresolved even after a strong earnings report from a single hyperscaler. As always, individual circumstances, time horizon, and risk tolerance should guide any specific investment decision, and this article should not be read as personalized financial advice.
What the past week has made clear, more than anything, is the degree to which the entire technology sector's valuation now hinges on a small handful of earnings prints from the largest cloud providers. A single confirmation that AI-related demand remains intact was enough to erase a multi-day correction in under 48 hours; a single disappointment, as investors saw with Alphabet's capex disclosure days earlier, was enough to trigger a swift repricing in the opposite direction. That sensitivity is likely to persist as the remaining hyperscalers, along with a broad swath of AI-adjacent chip and infrastructure companies, continue reporting results in the weeks ahead.
According to Bloomberg, CNBC, Reuters, Investing.com, Seeking Alpha, and 24/7 Wall St.