As the market prospects for PC continue to deteriorate, Intel is experiencing the worst stock price decline since the 1970s. The chip manufacturer is facing increasing pressure and must get its struggling wafer foundry business off the ground for good.

This has always been a challenge for Intel. Over the past five years, the company has invested a significant amount of resources and spent a considerable portion of its $108 billion in capital expenditures during this time attempting to compete with TSMC. However, it has yet to secure any iconic customers for its most advanced chips and currently mainly produces chips for its own products.

Wall Street's patience is waning. In August 2025, the U.S. government acquired a 10% stake in Intel, followed by NVIDIA's investment of $5 billion. Intel's stock price began to soar, rising sixfold over the 10 months before it peaked in June. Intel finally found its place in the wave of artificial intelligence, providing central processing units for AI, and at the same time, market optimism regarding its ambitions in wafer manufacturing was also on the rise.

But Intel's story took a dramatic turn for the worse on Thursday. The company stated that due to soaring costs of memory and other components, and the continued pressure on the PC market, revenue this quarter is expected to be $1 billion to $2 billion lower than what analysts anticipated. Intel still relies on its PC business for about half of its revenue, and consumers are becoming increasingly reluctant to upgrade their devices.

On Friday, Intel's stock price plummeted by 29%, marking its worst single-day performance since the 1970s, and also bringing its cumulative decline in October to 45%, making it the largest monthly decline on record. FactSet data shows that the stock fell another 8% during trading on Monday.

Intel's investments in the future were supposed to start showing results. CEO Chen Liwu stated that he expects the global server processor market size to grow to up to five times its current size over the next five years, reaching approximately 200 billion US dollars. Intel's data center business saw a 78% year-on-year increase in revenue in the third quarter, reaching 7.3 billion US dollars.

However, Intel stated that the company is unable to produce enough chips to meet the demand and pointed out that there are supply constraints across the industry for wafers, packaging substrates, and memory, which are also many of the same factors that are affecting the sales of PC.

This forced Chen Liwu to make a difficult choice. To address the shortage of production capacity, Intel stated that it is shifting some of the production capacity from its Oregon factory, which was previously used for manufacturing wafers for advanced 18A chips, to producing more wafers using a older process node Intel, which are used in its data center products.

Intel is hoping that the 18A will attract external wafer foundry customers, so such a shift in resources could complicate these efforts. David Yoffie, a professor at Harvard Business School who served on Intel's board of directors for nearly 30 years, stated that Intel's latest move is hardly likely to be well-received by potential foundry customers.

Yoffie said, "What they worry about is that if Intel actually doesn't have any extra production capacity beyond its own needs, then it won't be a very credible supplier." He said that if Intel cannot find a large-scale foundry customer within a year, "then the economic pressures on Intel will begin to become more apparent."

Intel did not respond to requests for comment.

Intel is best known for its industry-leading central processing units, namely CPU. In the early days of the AI craze, NVIDIA and AMD's graphics processors dominated the market as they were adept at handling the large-scale workloads required for model training and chatbot inference; whereas processors like Intel's CPU are crucial for virtual machines used in proxy applications such as Muse that run Meta.

Competitors or Partners?

Mizuho Securities analyst Jordan Klein stated in an interview that as one of the few companies in the world that can both manufacture and package the advanced chips required for their own products, Intel has "more control over its own destiny" than most chip manufacturers.

Klein said, "If you don't have sufficient production capacity, it's like leaving your revenue and profits on the table."

Pat Gelsinger, the former CEO of Intel who was removed from his position at the end of 2024, actively promoted investments in wafer foundry during his four-year tenure in order to catch up with the company's delays in 7-nanometer and 10-nanometer technologies. These advanced manufacturing processes are aimed at allowing each chip to accommodate more transistors.

Chen Liwu adopted a more cautious approach by limiting expenditures before customer needs were finalized. He stated on Thursday that he would announce "several customers" in the next two quarters.

Intel's own 18A chip has already achieved large-scale mass production in Arizona, and a second wafer factory is also under construction. Meanwhile, Intel's new factory in Ohio has just installed its first structural truss, although this project is nearly five years behind schedule compared to the original plan.

Yoffie indicates that another obstacle in acquiring new customers is that chip manufacturers such as Broadcom and Qualcomm may hesitate about whether to supply their chips to a company that is also a competitor at the product level.

Yoffie said, "The correct solution is to have Intel split its wafer foundry business." He claimed that this "is not only the best solution, but perhaps also the only solution."

Klein indicates that AMD might be the only large chip company not interested in manufacturing chips at Intel, as the two companies directly compete in the CPU market. He mentioned that this involves 'competition, confidentiality, and privacy issues.'

AMD will announce its third-quarter results after the market closes on Tuesday.

Yoffie indicates that the United States needs Intel to succeed.

He said, "In the end, TSMC can at most establish about 20% to 25% of its global production capacity in the United States, and it will still take several years to do so. If we don't have a domestic wafer foundry in the U.S. and encounter various geopolitical issues, or supply chain problems like those during the COVID-19 pandemic, that would be detrimental to this country as well as to the U.S. semiconductor industry."