Intel earnings signal recovery at US chip maker

Intel reported Q3 revenue of $13.6 billion, up 7% year over year, and its $3.7 billion loss came in better than expected, while it forecast current-quarter revenue of $13.8 billion to $14.8 billion. The results and AI-related commentary from CEO Lip-Bu Tan lifted shares more than 19% in after-hours trading, reinforcing optimism around Intel's turnaround. Management highlighted rising demand from AI inference and agentic computing for Intel CPUs, wafer, and advanced packaging offerings.

Analysis

The market is re-rating Intel less on a single print and more on a changed capital-allocation regime: once a foundry strategy gets explicit state and strategic-backer support, the downside is increasingly policy-protected while optionality on turnaround remains uncapped. That creates a very different equity profile from the last cycle—less like a pure execution story and more like a quasi-industrial policy asset where each incremental margin recovery matters disproportionately to valuation multiples.

The second-order winner is not Nvidia’s core GPU franchise, but adjacent infrastructure demand tied to inference, edge compute, and mixed workloads. If agentic AI shifts some spend away from centralized training clusters toward distributed CPU-heavy deployments, Intel can participate in a larger TAM without needing to win the highest-end AI silicon race. The loser is Taiwan-centric custom manufacturing leverage: any credible Intel foundry progress modestly reduces the monopoly rent embedded in advanced node capacity and gives hyperscalers another bargaining chip in sourcing negotiations.

The near-term risk is that the stock has already priced in a lot of good news, so the next 1-2 quarters may hinge less on earnings beats and more on whether gross margin and capex efficiency actually improve. If guidance disappoints, the market can quickly reframe this as a relief rally inside a still-fragile turnaround. Over a 6-18 month horizon, the key catalyst is foundry customer conversion; without visible external wafer wins, the narrative can stall even if product execution improves.

Consensus may be underestimating how much this changes competitive behavior across semis: cloud buyers now have stronger incentive to dual-source, and that can pressure pricing not just at Intel but across the ecosystem. The move is probably still directionally right, but the easy money in a straight long may be behind us after the post-earnings gap. Better risk/reward likely sits in relative-value expressions rather than outright chasing the upside.

Key Decisions for Investors

• Maintain a tactical long INTC, but only on post-gap consolidation or a pullback toward prior breakout levels; target a 3-6 month trade with upside tied to margin stabilization, not further multiple expansion.
• Pair trade: long INTC / short TSM on a 2-4 month horizon if the market starts pricing more credible US onshore foundry share; thesis is not TSM collapse, but modest multiple compression from incremental competitive pressure.
• Use NVDA weakness to add selectively only if the market misreads Intel strength as GPU substitution; core NVDA demand remains intact, so this is not an obvious short, but its valuation can be a hedge against AI-infrastructure enthusiasm overshoot.
• For event-driven accounts, buy INTC call spreads 3-6 months out rather than stock; defined risk is preferable because the name can mean-revert sharply if the next guidance print is merely in line.
• Avoid chasing TSM into the headline if the market extrapolates too much lost share; a better entry is on any follow-through selloff once investors price in slower, not absent, US foundry competition.