#IntelQ2RevenueSurges25%

Intel Just Pulled Off Its Best Quarter in 15 Years And the Story Beneath the Headlines Is Even More Interesting

$16.13B in revenue. 25% year-over-year growth. Non-GAAP EPS of $0.42 when Wall Street penciled in $0.21. Those aren't just beats they're demolition jobs.

Intel's Q2 2026 numbers landed Thursday after the bell and immediately rewired the narrative around a company that spent most of the past decade being written off as a legacy player trapped in its own complexity. Seven consecutive quarters of topping guidance. The strongest revenue growth since 2011. A stock that had already climbed 170% year-to-date before Thursday's after-hours surge pushed it another 13% higher toward $104.61 .

But here's what the headline percentages don't capture.

The engine isn't the PC it's the data center. DCAI revenue hit $6.26B, up 59% year-over-year. That's more than double the company's overall growth rate. AI agents are resurrecting CPU demand in ways nobody at Intel anticipated even six months ago. CEO Lip-Bu Tan admitted the shift caught them off guard demand is outrunning their ability to manufacture enough chips. That's a different kind of supply crisis than Intel's old problem of not having enough demand.

The foundry is alive. Intel Foundry posted $5.8B in revenue, up 31% nearly doubling the 16% growth it managed in Q1. 18A-P entered risk production. Panther Lake is in high-volume manufacturing using ASML's High-NA EUV technology, which puts Intel in territory few fab operators have reached. Fortinet just signed on as the first named external foundry customer under Tan's tenure . It's not TSMC-level volume yet, but the trajectory matters more than the current number.

The catch? Supply won't ease soon. Tan told analysts that the industry is facing "one of the most severe supply constraints in its history" across leading-edge logic, wafers, memory, and substrates. He cited contacts at major memory producers saying there's "no relief until 2028" . Intel raised its FY26 capex plan from $18B to $20B to accelerate investment in equipment, clean room space, and substrates a bet that demand will stay structural, not cyclical.

The accounting ghost underneath. GAAP net loss of $11B. EPS of $(2.16). That's a $12.5B non-cash charge tied to shares owed to the U.S. government under the CHIPS Act arrangement, not an operating disaster. Non-GAAP net income swung to $2.2B profit from a $441M loss a year ago. Operating margin jumped to 17.2% non-GAAP, up 21.1 percentage points YoY. Anyone reading only the GAAP line misses the actual operating story entirely.

Q3 guidance: $15.8B–$16.8B vs. $15.06B consensus. The midpoint implies a sequential decline from Q2's peak, which bears will note. But it still tops expectations, and EPS guidance of $0.38 vs. $0.27 consensus continues the trajectory.

The market's initial reaction was electric after-hours peaked near $113.72 before settling around $104.61. But Friday's regular session told a more complicated story: the stock closed down nearly 8% as the capex increase and foundry's still-heavy operating losses ($2.1B in the quarter) gave shorts ammunition . A company printing its best growth in 15 years still faces the tension between investing aggressively enough to capture AI demand and proving the foundry can eventually stand on its own economics.

Here's the framing that matters: Intel is no longer asking if AI demand is real. It's asking whether it can build fast enough to serve it. That's a fundamentally different question than the one this company was answering a year ago.