Continue Reading With Our 7-Day Free Trial
Intel reports second-quarter 2026 results after the market closes on Thursday, July 23, followed by a call at 5:00 p.m. ET. Management’s own guidance set the bar: revenue of $13.8 billion to $14.8 billion and adjusted earnings of $0.20 per share — both issued after a first quarter that beat expectations for a sixth straight time. The obvious debate is whether Intel can keep stringing together beats while its AI-driven CPU business scales.
But that’s not really the tension that matters this week. The more interesting question is whether the Intel 18A Process, Intel’s flagship manufacturing node, is progressing from something that merely works inside Intel’s own chips toward something outside customers will actually pay for. That distinction — technically functional versus commercially viable — is where this quarter’s real story lives, and it’s not something a single revenue number can settle.
There’s also a timing wrinkle. Some of the most consequential signals — 14A customer commitments, foundry breakeven — aren’t expected to fully show up this quarter at all. They’re a two-to-three-quarter story. Thursday is one data point in a longer arc built around the Intel 18A Process, not the finish line.
The Priced-In Scenario
Wall Street’s base case is built around the numbers Intel itself provided. At the midpoint of guidance, that’s roughly $14.3 billion in revenue with a non-GAAP gross margin near 39% — actually a step down from the 41% Intel posted in the first quarter, largely because Panther Lake, Intel’s product line built on the Intel 18A Process, was expected to roughly sextuple in volume quarter-over-quarter while still running below the corporate average margin.
The narrative Wall Street wants confirmed is straightforward: that CPUs are reasserting themselves as essential AI infrastructure, that data-center CPU demand keeps climbing as workloads shift from training toward inference and agentic use cases, and that Intel’s ASIC and advanced-packaging businesses keep compounding off a small base.
What that narrative tends to skip over is the foundry side of the house. External foundry revenue was still under $200 million last quarter — a rounding error next to the internal manufacturing business. The bull case implicitly assumes that gap closes on schedule. It’s a reasonable assumption, but it’s an assumption, and it only holds if the cost curve on the Intel 18A Process bends fast enough to keep pace with Panther Lake’s ballooning volume, because until that curve bends, every additional unit shipped adds revenue while still weighing down the margin line.
The Metric That Actually Matters
Set aside the top-line beat-or-miss game. The number that actually drives how this stock gets re-rated is the pace of yield improvement on the Intel 18A Process, and whether it’s translating into margins that look like the rest of Intel’s foundry business, rather than a persistent drag.
Here’s the mechanism. Intel Foundry currently absorbs the bulk of the cost of ramping a brand-new node. As long as yields on the Intel 18A Process sit below the threshold needed for healthy margins, every incremental Panther Lake shipment is a mixed blessing — good for revenue, a headwind for gross margin. Management has said yields are running ahead of internal targets, but “ahead of an internal target” and “at parity with mature nodes” are two different milestones, and the market has been pricing something closer to the second one.
This is also where the framing matters more than the headline. The fair test for Thursday isn’t whether Intel announces a marquee new foundry customer — it may not, and a quiet quarter on that front wouldn’t be a failure. The fairer test is whether the 18A-P variant’s move into risk production, and 14A’s development milestones, show forward motion toward external customers actually building products on Intel silicon — not just Intel building products on its own silicon.
Where Sentiment Could Turn Fast
There’s a second layer to this setup that has nothing to do with Intel’s own execution. Bank of America’s bubble-risk indicator for the semiconductor sector has been running elevated — a level that’s drawn comparisons, however loosely, to prior periods of stretched valuations across chip stocks broadly. That reading applies to the sector, not to Intel specifically, and it doesn’t mean sentiment is uniformly euphoric. But it does mean the whole group is trading with less room for error than it was earlier this year. A soft quarter from any major chip name right now risks a disproportionate reaction, simply because positioning across the sector is already stretched.
If Thursday Goes Right
The upside case doesn’t require a blowout number. It requires evidence that the yield curve on the Intel 18A Process is bending faster than the market assumed — visible progress on 18A-P’s readiness for outside customers, continued momentum in the 14A development timeline, and commentary suggesting design engagements are moving from “evaluating” to “committing.” Combine that with continued strength in data-center CPU demand and the ASIC business, and the psychological shift is significant: Intel stops being framed as a turnaround story that needs to be taken on faith and starts being framed as a foundry business with a visible, if still distant, path to relevance. In a sector this jumpy, a credibility upgrade like that can move the stock more than the quarter’s actual revenue delta.
If Thursday Goes Wrong
The downside case isn’t really about missing the revenue range — Intel has beaten its own guidance for six consecutive quarters, and few expect that streak to break cleanly. The real risk is a margin or commentary disappointment: gross margin guidance for the next quarter coming in soft, external foundry revenue staying negligible, or cost inflation in memory and substrates biting harder than flagged. Layered on top of an already-jittery semiconductor tape, even a modest disappointment could get amplified well beyond what the underlying numbers justify. Investors may be underestimating just how little cushion the current setup leaves for anything short of continued, visible progress on the Intel 18A Process.
Beyond This Quarter
The more important signals sit six to twelve months out. Management has pointed to the second half of 2026 through the first half of 2027 as the window when 14A design commitments should start becoming concrete. Intel Foundry’s break-even target is set for the exit of 2027, and the company’s long-term aspiration — a “Rule of 45” combining growth and profitability — is explicitly a multi-year goal, not a Thursday deliverable. Watch, too, how the competitive picture develops: AMD’s push on multithreaded server chips, and the growing presence of Arm-based alternatives from major cloud customers, all shape how much pricing power Intel’s CPU franchise can actually hold onto as the AI infrastructure buildout matures.
The Bottom Line
Thursday’s print will almost certainly generate a headline reaction — up or down — within minutes of the release. The more useful read comes from what’s said about the Intel 18A Process, external customer traction, and margin trajectory — not from where revenue lands inside an already-known range. That’s the part worth watching closely as the numbers come out, and worth revisiting once the call transcript is fully digested.
Disclaimer: We do not hold any positions in the above stock(s). Read our full disclaimer here.
