Things From Intel 10K That Make You Go …. Hmmmm

Key Takeaways

• Intel's manufacturing limited by Intel 7 node, not TSMC
• Gross margins ~35%; GAAP operating margin now negative
• Foundry loses money even with $7 B external revenue
• 2025 write‑offs due to inventory below market value
• New fabs profit split 50% with Apollo, Brookfield

Summary

Intel’s 2025 Form 10‑K reveals manufacturing constraints centered on its Intel 7 process rather than external partners like TSMC. Gross margins sit near 35 percent, yet GAAP operating margins turned negative after a large legacy asset write‑down and depreciation changes. The Foundry division would still lose money even with a hypothetical $7 billion external customer, and 2025 inventory write‑offs further pressure earnings. New fabs (34 and 52) involve profit‑sharing agreements with Apollo and Brookfield, adding financial risk if milestones are missed.

Pulse Analysis

Intel’s 2025 Form 10‑K makes clear that the company’s manufacturing bottleneck stems from the Intel 7 process, a node launched in 2021, rather than from external partners such as TSMC. While the older node can still meet certain legacy product demand, it limits the ability to scale advanced logic volumes needed for data‑center and AI workloads. Analysts interpret the constraint as a sign of strong demand, yet the reliance on an aging technology risks eroding Intel’s competitive edge in a market increasingly dominated by sub‑7 nm processes.

The financial section shows gross margins hovering around 35 percent, but GAAP operating margins have slipped into negative territory after a one‑time write‑down of legacy assets and a shift to a more aggressive depreciation schedule. These accounting moves saved billions in cash flow but also highlighted the underlying cost pressure from under‑utilized fabs and inventory that is recorded at the lower of cost or market. The 2025 write‑offs on 18A production underscore a mismatch between manufacturing capacity and market pricing, further compressing profitability.

The Foundry business remains a financial drag; even a hypothetical $7 billion external customer would not bring IFS to profitability, indicating structural economics issues. Moreover, Intel’s new fab projects—Fabs 34 and 52—are tied to profit‑sharing arrangements that give Apollo and Brookfield a 50 percent stake, creating additional cash‑flow obligations if milestones are missed. These dynamics force Intel to reconsider its IDM‑2.0 roadmap, potentially accelerating a shift toward niche, high‑margin products like Panther Lake while de‑emphasizing large‑scale foundry services. The market will watch closely whether the company can realign its cost base and regain investor confidence.