Intel Stock Falls 45% in October as Foundry Business Faces Credibility Test

Intel recorded its steepest monthly stock decline on record in October, falling 45% after a 29% single-day plunge on Friday — the company's worst one-day drop since the 1970s — as a deteriorating PC market and persistent foundry struggles rattled investors, according to data from FactSet.

The sell-off was triggered by Intel's warning that revenue this quarter would come in between $1 billion and $2 billion below analysts' estimates, with the company citing soaring costs for memory and other components as consumers increasingly opt not to upgrade their PCs. The stock fell an additional 8% in Monday trading.

Intel still depends on its PC segment for roughly half of its total revenue, leaving the company exposed as that market softens. CEO Lip-Bu Tan said he sees the global market for server processors growing as much as fivefold over the next five years to approximately $200 billion, but the company's ability to capitalize on that growth is now in question.

Intel's data center business posted 78% year-over-year growth in the third quarter, reaching $7.3 billion. But the company said it cannot produce enough chips to meet demand, pointing to industry-wide supply constraints on wafers, substrates, and memory — many of the same factors weighing on PC sales.

In response to the capacity shortage, Intel said it is redirecting some production capacity at its Oregon facilities away from its advanced 18A process node to manufacture chips on an older node known as Intel 3, which is used for its data center products. The 18A node is central to Intel's strategy for attracting outside foundry customers, and the reallocation could complicate that effort.

David Yoffie, a Harvard Business School professor who previously spent nearly three decades on Intel's board, said the shift sends the wrong signal to prospective customers. "They're concerned that if Intel doesn't actually have excess capacity beyond its own demand, that they're just not a very credible supplier," Yoffie said. If Intel fails to land a large-scale foundry customer within a year, "then the stress on Intel's economics are going to start to show even more dramatically," he added.

Tan said on Thursday that "a couple customers" would be announced over the coming two quarters. Intel's 18A chips for its own use are in high-volume production out of Arizona, with a second facility under construction. Intel's Ohio facility, meanwhile, is nearly five years behind schedule.

Jordan Klein, an analyst at Mizuho Securities, noted that Intel retains an advantage in controlling its own manufacturing. "If you don't have enough capacity, you leave revenue and money on the table," Klein said. He added that AMD is likely the only major chipmaker uninterested in manufacturing at Intel, given the two companies' direct competition in the CPU market.

Yoffie argued that Intel's foundry and product businesses may be structurally incompatible under one roof. Chipmakers like Broadcom and Qualcomm may hesitate to manufacture chips at a company that also competes against them on the product side. "The right solution is for Intel to spin off the foundry business," Yoffie said, calling it "not only the best solution, but maybe the only solution."

Intel's foundry ambitions carry national security dimensions as well. The U.S. government took a 10% stake in Intel in August 2025, and Nvidia followed with a $5 billion investment — moves that helped push the stock up sixfold over 10 months before it peaked in June. Yoffie cautioned that TSMC can only build "maybe 20-25% of their global capacity in the United States," and that would still take several years. "If we have no American foundry and we run into any number of geopolitical problems, supply chain problems like we had under Covid, that is going to be bad for the country and bad for the American semiconductor industry," he said.

Intel did not respond to a request for comment. Tan's decision to limit spending until customer demand is secured marks a deliberate break from his predecessor Pat Gelsinger, who aggressively pursued foundry investments during his four-year tenure before being ousted in late 2024. Whether that more restrained approach can produce a marquee customer fast enough to stabilize Intel's economics will be the defining question for the company heading into 2027.

Disclaimer