Earnings call transcript: Intel beats Q3 2026 estimates, but stock sinks

Intel shares fell 20.87% after hours to $72.85 despite Q3 adjusted EPS of $0.52 beating the $0.40 consensus and revenue of $16.5B slightly topping estimates. Investors focused on Q4 guidance of $15B-$16B revenue and about $0.25 adjusted EPS, plus an estimated $300M revenue hit and roughly 200bp gross-margin pressure from the Israel plant shutdown. Data center and AI revenue rose 17% sequentially to $7.3B, while client computing revenue fell 9% sequentially to $8.1B.

Analysis

The key issue is conversion of demand into shippable units, not whether server CPUs have a compelling AI use case. Intel’s wafer and advanced-substrate constraints leave room for AMD and Arm-based platforms to capture customer demand; reallocating capacity away from PCs may help, but it does not remove the substrate bottleneck. Oracle’s deployment agreement is a positive demand signal, not evidence that Intel can yet deliver at scale.

The earnings beat is lower quality than the headline implies: operating execution improved, but the next-quarter guide resets earnings and margins sharply lower, and management explicitly expects no positive free cash flow in 2027. The equity raise reduces near-term funding risk while diluting per-share recovery potential. Foundry progress is also easy to overread: the sequential increase in external revenue was primarily tied to a non-recurring program, while the business remains loss-making. Treat customer interest and yield claims as milestones to verify, not yet as proof of durable external economics.

Near term, the after-hours gap creates squeeze and price-discovery risk; avoid chasing a short at the open. Over 1–3 months, watch the Q4/Q1 disruption, client inventory correction, and substrate availability. Over 6–18 months, the thesis turns on whether Intel converts improving yields and higher investment into market-share gains and cash returns. The contrarian upside is that server CPU scarcity supports a larger market and Intel can recover if supply loosens; the market may be discounting that option, but 2027 cash consumption raises the execution hurdle.

Key Decisions for Investors

• Tactical relative-value watch: consider long AMD / short INTC only if regular-session trading confirms persistent INTC underperformance after the gap; keep the position small and do not chase an opening dislocation. The thesis is rival share capture while Intel is supply-constrained. Falsify it if Intel demonstrates improving server shipments and substrate availability without further guide reductions.
• For INTC, wait for price discovery rather than buying solely on the EPS beat. Reassess after Q4 results and the January update: evidence of Intel 3/18A yield improvement is insufficient unless DCAI shipments rise and gross-margin pressure stabilizes. A further disruption or weaker first-half 2027 supply outlook would reinforce downside risk.
• Track PC memory pricing and availability as a key cross-sector signal. If shortages persist, treat Intel’s client weakness as potentially structural into 2027—not merely a short inventory flush—and monitor whether demand shifts toward lower-cost systems or competing CPU architectures.
• Treat Oracle’s announced CPU deployment as a watch item, not a standalone ORCL trade: verify delivery timing and scale. For Intel, monitor external foundry revenue excluding one-off programs, foundry operating losses, and capital spending against cash generation before underwriting a foundry rerating.