Intel’s Make-or-Break Earnings: After a 4-Year Breakout, Is $48 the Trapdoor?

Intel reports after the close tonight, and the setup is about as clean as it gets: the stock has been ripping, the narrative has shifted from “can they survive?” to “can they execute flawlessly?”, and the valuation is now doing the thing that valuations do when investors get religion. Shares broke out above $48 yesterday and pushed to a four-year high, which effectively turns this earnings print into a referendum on whether the turnaround story deserves a new multiple—or whether the market got ahead of itself and needs to re-test that breakout level as support.

The overarching is a tug-of-war between improving demand in Intel’s core CPU franchises and the long, expensive climb back to manufacturing credibility. Bulls are leaning into three ideas: (1) traditional server CPU demand is tightening as AI workloads pull through broader data-center buildouts; (2) Intel’s PC roadmap (Panther Lake) is arriving at a moment when OEMs and enterprises actually want a refresh cycle; and (3) the foundry narrative—18A today, 14A later—can re-rate Intel from a cyclical product company into a strategic U.S.-based manufacturing platform. Bears aren’t arguing those themes are imaginary; they’re arguing Intel’s current product competitiveness and margin structure don’t justify the speed (and size) of the re-rating.

That valuation tension is visible when you stack Intel against peers. On earnings multiples, Intel screens “expensive/awkward” because the earnings base is still impaired: the forward non-GAAP P/E is roughly 159x , compared with ~39x for Nvidia, ~25x for TSMC, and ~32x for Broadcom (AMD is also rich at ~63x FY1). On trailing results it’s even messier: Intel’s non-GAAP TTM is effectively not meaningful and GAAP TTM is around 1,205x—basically telling you the market is paying for a future earnings recovery that hasn’t shown up cleanly yet. On sales multiples, though, Intel looks cheaper: ~4.5x price-to-sales and ~5.5x EV/sales, far below Nvidia and Broadcom (~24x-ish) and below AMD/TSM (~11–13x). That’s why the stock can look “value-like” to one investor and “bubble-ish” to another depending on which denominator they trust. The real debate is whether the earnings denominator starts catching up.

For the quarter itself, expectations are modest and that matters because the stock isn’t trading like expectations are modest. Street numbers clustered around adjusted EPS of about $0.08–$0.09 on revenue around $13.4B for Q4, down versus the year-ago period. That’s the baseline. What investors will actually trade is guidance, margins, and any evidence Intel is converting demand into sustainable profitability rather than simply operating in a constrained supply environment. Morgan Stanley’s framing is a good representation of the skeptical camp: they acknowledge CPU shortages can support the turnaround, and they’ve seen server tightness for months, but they question whether the longer-term earnings power supports the re-rating. They also argue that a meaningful chunk of incremental server demand may be flowing to AMD, which puts a ceiling on how much Intel can “win” even in a rising tide.

The first “tell” in the print is going to be the product segments and what they imply about share. Investors will look for signals that demand is strong in both Client and Data Center, but also whether Intel is under-shipping demand due to capacity constraints—and if so, whether that’s actually translating to pricing power or just pushing customers to alternative supply. Commentary around Granite Rapids upgrades, backlog, and supply visibility into 2026 will be key. If management describes the server market as effectively sold-out but can’t show meaningful operating leverage, the market will start asking uncomfortable questions about mix, yields, and competitiveness.

The second “tell” is gross margin, because margin is where narratives either become businesses or remain PowerPoint. There’s already a widely-circulated view that gross margins are under pressure due to mix, early ramps, and the cost of building out leading-edge manufacturing. If Intel prints or guides to margins that look stubbornly low despite demand strength, it reinforces the bear case that this is a supply-constrained, cost-heavy recovery—especially if AMD continues to capture the bulk of the incremental server growth.

Capex is the third—and maybe most important—swing factor. For Intel, capex isn’t just a spending line; it’s the credibility line. Foundry progress requires enormous, sustained investment in tools, process development, and capacity, and management has repeatedly framed capex discipline as essential to protecting liquidity. Investors will want a clear capex number/trajectory and a rationale that ties spending to (a) committed demand, (b) yield improvements, and (c) the cadence of 18A ramp plus the longer-dated 14A push. If capex comes in higher, the market will ask whether Intel is chasing TSMC’s scale without TSMC’s customer pull-through; if capex comes in lower, the market will ask if Intel is under-investing just as the AI-driven manufacturing arms race accelerates. Either way, capex is the fulcrum because it determines whether “foundry optionality” is a real call option or just an expensive hobby.

That brings us to Intel Foundry Services (IFS), which is doing heavy lifting in the stock even though it’s still a financial drag. Investors want two things here: proof that 18A is real (performance, yields, schedule) and evidence of external customer momentum (even small volume wins, packaging deals, qualification milestones). The bullish camp, reflected in Seaport’s upgrade, is leaning on strong signals from PC OEMs/ODMs for Panther Lake and the idea that the commercial launch of 18A marks Intel’s return to a credible process-improvement path. They also make a practical point: even if 18A alone won’t “save” IFS, Intel can get traction via advanced packaging while TSMC’s CoWoS capacity is constrained. That’s an underappreciated wedge—packaging wins can be a beachhead.

The skeptical camp is essentially saying: it’s hard to sell yourself as the manufacturing alternative when your own products are struggling to fully capitalize on demand. Morgan Stanley’s note captures that: foundry narratives may be bolstering the stock, but customers evaluating Intel Foundry will notice if Intel can’t meet demand in its own product lines or if product competitiveness lags for multiple years. That’s why the market keeps circling back to the same catalysts: Panther Lake reception, 18A validation, and any “real partner” announcements versus rumor mill chatter (Apple is the obvious name everyone whispers).

The political overlay has also become part of the story, and it’s not subtle. President Trump’s January 8 post poured gasoline on the “strategic national asset” framing, writing: “I just finished a great meeting with the very successful Intel CEO, Lip-Bu Tan… Intel just launched the first SUB 2 NANOMETER CPU PROCESSOR designed, built, and packaged right here in the U.S.A… Our Country is determined to bring leading edge Chip Manufacturing back to America, and that is exactly what is happening!!!” In market terms, that kind of endorsement reinforces the perception of policy tailwinds and potentially supportive capital, which can matter a lot when you’re funding a multi-year manufacturing rebuild.

So what happens if results falter? Technically, the breakout above $48 now matters. If the company disappoints on guide, margins, or credibility markers (capex clarity, foundry milestones, customer traction), a pullback that “tests the breakout” is the cleanest market response. A dip toward $48 would effectively ask: was that move real accumulation ahead of fundamentals, or a narrative squeeze that needs to reset? Conversely, if Intel delivers even a modest beat/raise with improving margin commentary, clearer capex discipline, and incremental foundry proof points, the market can justify staying long the re-rate—even if near-term EPS still looks unimpressive.

Tonight’s report is less about whether Intel prints $0.08 or $0.09, and more about whether management can connect the dots between demand, execution, and a credible path to earnings power. The stock is priced like those dots are already connected. Now Intel has to show its work.