Intel's earnings report on July 23 should have been a victory lap. Instead, it turned into a case study in how a single line of guidance can undo an otherwise flawless quarter. Shares jumped as much as 13% in after-hours trading the night of the release, briefly touching above $112. By the close of trading two days later, on July 25, the stock had reversed entirely, finishing at $92.32, down 7.9% on the day and erasing roughly $40 billion in market value based on Intel's July 17 share count.

A quarter that beat on almost everything

The headline numbers were hard to argue with. Revenue came in at $16.13 billion, beating Wall Street's consensus estimate of roughly $14.43 billion by about 12%. Non-GAAP earnings per share landed at $0.42, nearly double the $0.21 to $0.22 analysts had modeled. According to CEO Lip-Bu Tan, the quarter marked Intel's seventh consecutive period of exceeding its own financial guidance, a streak that has quietly rebuilt investor confidence in a turnaround story that spent much of the past two years running on faith rather than results.

Non-GAAP gross margin reached 41.8%, beating the company's own guidance by 280 basis points, driven by higher average selling prices, improved manufacturing yields, and a more favorable product mix. CFO Dave Zinsner credited the beat to "robust demand and improved execution, including volume upside driven by higher factory yields and improved cycle times." Non-GAAP operating margin improved to 17.2%, a dramatic swing from negative 3.9% in the same quarter a year earlier, even as the company reported a GAAP net loss of $11 billion, driven primarily by non-cash charges rather than operating weakness.

The foundry business shows real, if modest, progress

Intel's foundry division, the unit manufacturing chips for outside customers and central to Intel's long-term turnaround thesis, generated $5.8 billion in revenue, up 31% year over year and roughly 6% sequentially, though the segment still posted an operating loss of $2.1 billion. That loss narrowed by $348 million from the prior quarter, a trend line investors have been watching closely for signs the foundry business can eventually break even.

External customer revenue, the portion of foundry business coming from companies other than Intel itself, reached $293 million for the quarter, up roughly 68% sequentially from $174 million in the first quarter of 2026. That is still a small fraction of the segment's total $5.8 billion in revenue, but reports that Google has placed an order with Intel to manufacture 3 million of its custom Tensor Processing Units, and that Nvidia is reportedly evaluating Intel as a manufacturing option, suggest the external customer pipeline may be building faster than the current revenue figures show. Analysts are watching for confirmation of Intel's first external customer commitment on its advanced 14A manufacturing process, expected in the second half of 2026, as the clearest signal yet of whether outside chipmakers are willing to bet flagship designs on Intel rather than incumbent foundry leader Taiwan Semiconductor Manufacturing Company.

Where the good news turned into a problem

The reversal traces directly to one number: capital expenditure guidance. Intel raised its full-year 2026 capex forecast to more than $20 billion, up roughly $3 billion, or about 17%, from its prior guidance of $18 billion. Management framed the increase as a sign of strength, tied to what Zinsner described as continued strengthening in AI-driven compute demand, requiring "meaningfully increasing our investments in equipment, clean room space, and substrates" to support growth across both product and foundry segments.

Investors read the same number differently. A capex increase this large, layered on top of a foundry segment still losing more than $2 billion per quarter, revived a concern that has shadowed Intel's entire turnaround: whether the company's massive, multi-year investment cycle can generate returns fast enough to justify its cost. That concern was compounded by reports that Intel may be reconsidering the scale of its Ohio manufacturing campus, with SK Hynix reportedly exploring a potential acquisition of part of that site, a development some analysts read as a signal that Intel's fab ambitions may be getting scaled back even as headline capex spending rises elsewhere.

A valuation that leaves little room for doubt

Context matters here as much as the numbers themselves. Intel entered its earnings report trading at roughly 94 times forward earnings, a valuation far above semiconductor peers such as Nvidia, Broadcom, Taiwan Semiconductor, and Micron, which trade between roughly 13 and 33 times expected earnings. That premium reflected enormous investor confidence that Intel's margin recovery and foundry turnaround were durable, following a run that had taken the stock up more than 300% over the prior year. A valuation stretched that far leaves almost no room for a mixed signal, and a large capex increase paired with continued foundry losses was evidently enough to trigger one.

Intel's own longer-term earnings estimates tell a sobering parallel story. Projected 2026 earnings per share have fallen from roughly $6 several years ago to just $1.11 today, while 2027 estimates have dropped from above $4 to below $2, underscoring just how much of Intel's current valuation rests on a recovery that is still, by the company's own admission, several years from completion.

A pattern showing up across the chip sector

Intel's swing from a 13% pop to a near 8% decline within 48 hours echoes a pattern that has repeated across the semiconductor sector in 2026. Broadcom, for example, posted stronger-than-expected fiscal second-quarter earnings in June, beating adjusted EPS estimates of $2.39 with a result of $2.44, only to see shares fall nearly 13% after-hours on disappointing forward guidance tied to AI infrastructure spending. The common thread across both companies is a market that has grown increasingly skeptical of capital expenditure increases specifically, treating them less as evidence of confidence and more as a warning sign about near-term profitability, regardless of how strong the underlying quarter actually was.

What the next few quarters need to show

Intel's third-quarter guidance, a midpoint of roughly $15.8 billion to $16.8 billion, sits above analyst consensus even though it represents a sequential decline from the second quarter's $16.13 billion peak. Whether that guidance holds, and whether the promised external foundry customers actually materialize on the 14A process later this year, will likely determine if Intel's current valuation premium survives or continues to compress. For a company whose turnaround narrative depends on outside investors trusting a multi-year, multi-billion-dollar bet on advanced manufacturing capacity, this week's reaction suggests that trust, while real, remains conditional on tangible proof rather than guidance alone.

---

*Sources cited in this article include Intel's official Q2 2026 earnings release and CFO commentary, and reporting from Yahoo Finance, Investing.com, Seeking Alpha, TS2.tech, and Money365.Market covering the July 23–25, 2026 earnings release and subsequent stock decline. All figures reflect reporting available as of July 27, 2026.*