Intel's radical proposal: "Split off the market" foundry and design departments. Can Trump and Nvidia join forces to reshape American chip hegemony?
The U.S. government and Nvidia have acquired nearly 20% of the shares. The company’s former director suggested privatization and splitting into scalpels, paving the way for Intel to get rid of the group’s discount and return to process leadership. The action affects the U.S. semiconductor security and AI supply chain, but it also faces huge capital, regulations and coordination problems.
(Previous summary: Nvidia announced a US$5 billion investment in Intel: joint cooperation in chip design, Intel surged 30% before the market opened)
(Background supplement: It’s not just Intel! The Trump administration’s “subsidy swap” has expanded to TSMC and Samsung, and national capital is coming fiercely)
Contents of this article
Intel Intel is at a turning point in fate: the US government holds nearly 10% Equity, Nvidia invests another $5 billion. According to the latest report from "Fortune", four former Intel directors called for a radical proposal to privatize and split the company. This move is not only a corporate governance issue, but also closely connected with geopolitics, supply chain security and AI competitiveness.
Why do the national team and leading players invest in shares?
Intel used to rely on vertical integration to dominate, but with the rise of competitors such as TSMC and AMD, the company's operations have declined. Four former directors called for a fundamental restructuring to return to competition.
They advocate delisting Intel through a consortium acquisition and then dividing the company into two: an independent foundry entity and a business unit focused on design. The foundry side will focus on serving global customers and will invest US$100 billion in the next ten years to catch up with TSMC; the design side will focus on PCs, servers and data center processors to release valuations that have been damaged by manufacturing.
On the other hand, privatization allows management to adjust the compensation structure, attract top engineers, and use relisting as a long-term inducement. In the future, we will refer to the experience of AT&T in the 1980s. The entire restructuring is expected to be completed within one year, and it can be listed or sold in batches before 2028.
Coordination challenges and potential resistance
However, the US CHIPS Act requires Intel to maintain a 51% stake in foundry entities for at least five years. According to Windows Central, this condition may limit a complete split. Furthermore, Intel's foundry has been losing billions of dollars every year in recent years, and its huge capital expenditures will test the consortium's patience.
It is equally complex at the operational level. Intel's internal design, process and packaging are closely intertwined, and cutting affects the supply chain, customers and manpower. Before the risks are clear, whether major customers such as Microsoft and Apple are willing to transfer orders to new foundry companies that have not yet been proven is also a big question mark.
In general, Intel’s management, board of directors, and investors currently have different views on this radical plan, and the regulatory approval process is also full of variables.
The next step for U.S. semiconductor hegemony
However, if Intel is indeed successfully broken up, the United States may improve the resilience of the AI supply chain and create high-paying jobs and tax revenue; but conversely, if capital or coordination fails, Intel may fall into a long-term transformation stagnation, making advanced manufacturing processes more dependent on overseas.
No matter how it ends, Intel's restructuring has transcended corporate success or failure and directly reflects the United States' position in the next generation technology race. The government, large technology companies and the capital market are jointly betting that whether the heavy burden on the profit and loss statement can be turned into strategic assets will become one of the most watched projects in the AI era.
