Intel shares are experiencing their most severe period in decades: by the end of October 2026, the company's quotes plummeted by 45%, marking the largest monthly decline in the history of observations for Intel securities. Friday's 29% crash — the worst single-day performance since the 1970s — was the logical conclusion of a series of negative signals that began with a revision of the quarterly revenue forecast. The company warned that its revenues in the current quarter would come in $1–2 billion below analyst expectations due to a deepening crisis in the personal computer market, rising costs for memory modules, and other components. Meanwhile, the PC segment still generates about half of Intel's total revenue, and end consumers are upgrading their devices less and less, turning consumer electronics dependency into a systemic risk.
$108 Billion Without External Customers
Intel's fundamental problem extends far beyond a single quarterly report. Over the past five years, the company has poured $108 billion in capital expenditures, predominantly into attempts to compete with Taiwan's TSMC in the advanced semiconductor foundry market. However, by November 2026, Intel still does not have a single major external customer for chip manufacturing on advanced process nodes: the company continues to produce semiconductors primarily for its own needs. This fact makes the strategic bet on the foundry business extremely vulnerable — production facilities are built and operated, but they do not generate the external order flow that was supposed to pay off the colossal investments.
From Sixfold Growth to Crash: The Role of Government and Nvidia
As recently as August 2025, the situation looked fundamentally different. The U.S. government acquired a 10% stake in Intel, and Nvidia invested $5 billion in the company. Against this backdrop, Intel quotes surged sixfold over ten months and reached an all-time high in June 2026. The market interpreted these events as confirmation that Intel had found its niche in the artificial intelligence ecosystem — specifically by establishing the supply of central processing units for AI agent systems. An additional boost came from a share offering: according to industry publications, Intel raised $20 billion, setting the offering price at $95 per share, which allowed it to finance the expansion of production capacities. However, the revision of the quarterly forecast and the admission of an inability to meet existing demand for chips shattered this narrative in a matter of days.
Resource Redirection in Oregon: The Price of Shortage
Amid industry-wide shortages of silicon wafers, substrates, and memory modules, Intel management decided to redirect some production resources at its Oregon factories from the advanced 18A process to the older Intel 3 technology. Formally, the company explains this by the need to ensure current shipments, but analysts point out the paradox: the 18A process was a key argument for attracting external customers, and its partial "shutdown" in Oregon could complicate the already difficult task of winning client trust. Intel itself assures that it is "following the technological roadmap, demonstrating excellent metrics within the 18A process and progress in developing 14A," and emphasizes high interest from potential customers. At the VLSI symposium in Honolulu, the company also announced the start of pilot production of chips using an improved version of the 18A-P technology.
AI Agents and the Server Market: Window of Opportunity
Despite the short-term crisis, Intel's long-term prospects in the data segment remain significant. CEO Lip-Bu Tan forecasts a fivefold growth in the global server processor market to $200 billion over the next five years. In the third quarter, Intel's data center solutions division showed a 78% year-over-year growth, reaching $7.8 billion in revenue. The structural shift in the AI ecosystem works in Intel's favor: while Nvidia and AMD graphics processors dominate model training and chatbot operations, AI agents require powerful central processing units, and Intel remains one of the few companies in the world capable of independently manufacturing and packaging advanced chips for such tasks. Mass production using the 18A process has already been established in Arizona, and preparations are underway to launch a second plant.
Competitive Pressure and Tan's Strategy
The key obstacle to implementing Intel's foundry strategy remains competition. Potential customers — Broadcom, Qualcomm — may refrain from cooperation due to fears of transferring technologies to competitors. AMD, as a direct competitor in the processor segment, will highly likely not become an Intel client; for it, this would touch upon issues of not only competition but also the protection of commercial secrets. Against this backdrop, TSMC, according to DigiTimes, plans to start mass production using the A14 (1.4 nm) process in the first half of 2028, with its production capacities potentially exceeding Intel's figures for the competing 14A process by 7.5 times by that time. Lip-Bu Tan, who replaced Pat Gelsinger in 2024, has taken a more cautious stance, capping spending until customer demand is guaranteed. Last week, he promised that Intel would announce cooperation with several customers in the next two quarters. Whether the company can convert its existing production capacities and technological advantages into real contracts before TSMC solidifies its position on the next process node is the main question determining Intel's future in the semiconductor market.
