On November 21, 2025, global manufacturing giant Foxconn (Hon Hai Precision Industry Co., Ltd.) announced a significant plan: to initially invest between $1 billion and $5 billion to expand its manufacturing footprint in the United States, while also establishing a deep collaboration with AI industry leader OpenAI for the design and manufacturing of AI infrastructure hardware. This series of actions not only marks Foxconn’s formal entry into the U.S. artificial intelligence sector but also highlights its strategic determination to reshape its business landscape and seize a leading position in the AI hardware ecosystem.
The collaboration with OpenAI precisely targets the core pain points of AI infrastructure construction, defining three main tasks. First, the two parties will jointly design and develop future generations of AI data center racks, covering the entire process of hardware design, engineering, and development, specifically addressing critical issues such as equipment failures during the startup and operation of AI data centers, thereby laying a solid hardware foundation for stable AI computing power output. Second, to adapt to local manufacturing needs in the U.S., the collaboration will optimize rack architecture, expand the scope of local procurement, and introduce more domestic chips and suppliers, helping to simplify the U.S. AI supply chain system and enhance supply chain security and responsiveness. Third, Foxconn will produce key components for AI data centers, such as cables, networks, cooling, and power systems, on local production lines in the U.S., achieving a localized production loop for core hardware. Foxconn Chairman Liu Yangwei provided clear capacity expectations: by 2026, the company could assemble up to 2,000 server racks per week, doubling its current capacity, demonstrating its rapid execution capability.
Behind this short-term large investment lies Foxconn’s longer-term strategic planning. Liu Yangwei revealed that over the next 3-5 years, Foxconn plans to invest $2-3 billion annually in the field of artificial intelligence, which will account for more than half of the company’s annual capital expenditure of approximately $5 billion. The continuous and high-intensity capital injection dispels external speculation about its “short-term trial” and highlights that the AI business has become a core growth driver for Foxconn’s future development.
Foxconn’s layout in the U.S. AI sector is not starting from scratch but is built on a solid local foundation. Currently, Foxconn has established factories in several locations in the U.S., including Wisconsin, Ohio, and Texas, forming a relatively complete manufacturing network. In May of this year, it was approved to invest $735 million to establish Project ETA (DE) LLC, which focuses specifically on data center equipment manufacturing and AI server assembly, providing a ready-made capacity carrier, technical team, and industrial chain resources for this large-scale expansion, significantly reducing the cost and risk of strategic implementation.
From a corporate development strategy perspective, entering the U.S. AI sector holds multiple profound meanings for Foxconn. First, it is a crucial step in accelerating its business transformation. For a long time, Foxconn has been highly dependent on Apple’s iPhone assembly business, and with the recent slowdown in the consumer electronics market, business transformation has become urgent. Its Q3 2025 financial report shows that revenue from cloud and networking businesses has exceeded that from consumer electronics for two consecutive quarters, becoming the company’s largest source of income. The investment in the U.S. AI sector will undoubtedly inject a strong impetus into this business segment, accelerating the transformation process and building a more balanced business structure. Second, this move precisely aligns with U.S. market demands, meeting the government’s requirements for localizing AI infrastructure while effectively reducing tariff risks in international trade, creating a more stable environment for business operations. Furthermore, by deeply binding with leading industry players, Foxconn will further consolidate its industry position. As a major manufacturer of NVIDIA AI racks, it has already taken on nearly 30% of NVIDIA’s AI server orders in 2024. The collaboration with OpenAI represents an upgrade from being a “hardware supplier” to a “core partner,” which will continuously strengthen its core competitiveness in the global AI hardware supply chain.
It is noteworthy that Foxconn’s U.S. AI layout is also forming synergistic effects. On the same day it announced its collaboration with OpenAI, Foxconn also revealed the establishment of a joint venture with Intrinsic, a subsidiary of Google’s parent company Alphabet, to explore robotics technology related to smart factories in the U.S. Robotics technology, as an important application direction under the AI boom, complements AI server manufacturing and data center construction, helping Foxconn build an industrial ecosystem of “AI hardware manufacturing + intelligent application implementation,” creating a strategic effect of 1+1>2.
From a consumer electronics contract manufacturer to an AI hardware ecosystem builder, Foxconn’s transformation path is clearly visible. This multi-billion dollar investment in the U.S. AI sector is not only a precise grasp of the global AI industry wave but also a proactive transformation to reshape the company’s core competitiveness. As production capacity gradually releases and collaborations deepen, Foxconn is expected to occupy a more central position in the global AI hardware supply chain, and its layout in the U.S. market will also bring new variables to the development pattern of the global AI industry.