Intel blows past estimates, recording fastest sales growth in almost 15 years on 'unprecedented' demand

Intel reported Q2 adjusted EPS of 42 cents vs $0.21 expected and revenue of $16.1B vs $14.42B, delivering 25% YoY growth (fastest since Q3 2011). For Q3, it guided adjusted EPS to 38 cents and revenue of $15.8B–$16.8B, both above LSEG expectations (revenue $15.1B, EPS 27 cents). Shares jumped ~11% in extended trading as AI-driven demand lifted data center sales 59% to $6.3B and gross margin recovered to 42%.

Analysis

The immediate read-through is not just a beat, but evidence that Intel’s existing CPU franchise is regaining operating leverage faster than the market expected. If server demand is still outstripping supply, the key consequence is pricing discipline: that can hold gross margin above the level investors had implicitly modeled for a slow-turn turnaround, which supports further multiple expansion over the next 1-3 quarters.

The more interesting second-order effect is competitive, not promotional. Intel using long-term customer agreements suggests enterprise buyers are willing to trade flexibility for supply assurance, which can slow switching away from x86 and reduce AMD’s near-term share-taking optionality. But the foundry story remains unvalidated: a named older-node customer is useful for utilization, yet it does not prove that Intel can win meaningful leading-edge external volume, so the capex step-up still looks like a cash-flow drag before it becomes a growth driver.

Risk is front-loaded in the next 1-2 earnings cycles: if data center growth decelerates, memory shortages ease, or margins slip back toward the mid-30s, the stock can give back a meaningful portion of the post-print move. Over 6-18 months, the thesis only compounds if Intel can disclose a real external foundry anchor and keep capex from outrunning free cash flow; absent that, the market may be paying for a turnaround narrative that is still only partially monetized. The contrarian view is that the crowd is likely underestimating how much of the rerating is already powered by the captive CPU business, while overestimating the near-term foundry option value.

Key Decisions for Investors

• Tactically buy INTC only on a post-earnings pullback over the next 2-5 sessions; use a 1-3 month horizon and size for continuation in estimate revisions, not a straight-line rerate. Falsifier: next-quarter revenue guide slips back below consensus or gross margin fails to hold above the low-40s.
• Express the relative-value view as long INTC / short AMD for 1-3 months if you want to monetize server-CPU share-defense and supply-constrained pricing. This works best if Intel prints another sequentially firm data-center update; it breaks if AMD re-accelerates AI-server share or Intel’s CPU growth normalizes quickly.
• Watchlist rather than immediate trade: long AMAT or LRCX into any confirmed Intel capex increase next year. The cleaner winner from Intel’s foundry spend is the equipment complex, but only if Intel keeps raising capex without cutting the plan back; otherwise the trade becomes a value trap.
• Do not chase FTNT on the foundry-customer headline alone. The initial node is not strategic validation of Intel Foundry, so the read-through is mostly noise unless FTNT or Intel discloses follow-on volume on a more advanced process.