Intel Corporation's (INTC.US) manufacturing strategy accelerates: Countdown to mass production of 18A.

At the Intel Foundry Direct Connect conference hosted by Intel Corporation (INTC.US) in 2025, Intel Corporation announced the latest progress of its "four-year five-node process" plan, marking a critical implementation phase of Intel Corporation's foundry strategy. CEO of Intel Corporation, Lip-Bu Tan, shared the progress and future development focus of Intel Corporation's foundry in the opening speech, emphasizing that Intel Corporation is advancing its foundry strategy to the next stage. Intel Corporation's foundry services also include mature nodes, with the first 16nm production wafers already completed in its fabs, and Intel Corporation is also attracting customers with the 12nm nodes developed in collaboration with United Microelectronics Corporation (UMC). However, the main focus of Intel Corporation's foundry services is on advancing advanced process nodes. Currently, the Intel 18A process node has entered the risk production stage with continuously optimized defect density. Intel Corporation plans to achieve large-scale production by the end of 2025. This node uses Power Via backside power supply technology, with a 20% increase in transistor density compared to the previous generation, and a performance improvement of 10%-15%. It is mainly targeted at high-end markets such as AI chips, High Performance Computing (HPC), and domain controllers for autonomous driving. As the core product of Intel Corporation's foundry strategy, the mass production of Intel 18A will directly compete with Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR's 2nm (N2) and Samsung's SF2 nodes. Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR's N2 process node is expected to enter mass production in the second half of 2025, while Samsung's SF2 node is planned for release in 2025. With the technological advantage of the 18A node, Intel Corporation has attracted top customers such as NVIDIA Corporation, Broadcom Inc., and MediaTek to conduct wafer tests. NVIDIA Corporation plans to develop the next generation AI accelerator card based on the 18A process. In the next generation process layout, Intel Corporation announced that Intel 14A (1.4nm) has started customer collaboration, sending early PDKs (Process Design Kits) containing a set of data, documents, and design rules for processor design and validation. According to Intel Corporation, the 14A process is expected to enter mass production in 2027, a year earlier than the Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR's A14 process (expected to enter production in 2028). Intel 14A is the next generation product after 18A, and if all goes well, 14A will be the industry's first node to adopt high numerical aperture EUV lithography technology. The 14A process technology will use Power Direct direct contact power supply technology, combined with the second-generation Ribbon FET transistor architecture, to achieve a performance improvement of 15%-20% per watt. In addition, Intel 14A will also introduce new Turbo Cell technology, allowing designers to dynamically optimize performance and power consumption balance within the chip module to further increase chip speed, including CPU maximum frequency and GPU critical paths, strengthening Intel Corporation's competitiveness in the AI and HPC fields. Intel Corporation stated in a press release that "Turbo Cells achieve a balance between power consumption, performance, and area for target applications." Intel Corporation emphasized that its second High NA EUV equipment has been put into operation for Intel 14A. This equipment starts much faster than the first equipment. Facing competition from Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR's SoIC and Samsung's X-Cube, Intel Corporation is building system-level integration capabilities through Foveros Direct 3D stacking and EMIB 2.5D bridge technology. At the Intel Foundry Direct Connect conference, Intel Corporation showcased the super-large-scale Chiplet solution based on EMIB-T technology, which can integrate 4 computing dies and 12 HBM5 memories, supporting a package size of 120mm120mm, to meet the high-bandwidth memory requirements of AI chips. In addition, Foveros-R and Foveros-B technologies will enter mass production in 2027, further enhancing packaging flexibility and energy efficiency. In terms of supply chain collaboration, Intel Corporation has partnered with Amkor Technology to incorporate Amkor's packaging capacity into the foundry ecosystem, providing customers with more packaging technology options. This "process + packaging" collaborative mode forms Intel Corporation's full chain service capability from chip design to system integration for customers, such as providing comprehensive solutions for NVIDIA C.The AI chip customized by Corporation can achieve 3D stacking through Foveros Direct, increasing computing density by more than 30%.orporation Through a diversified global manufacturing network, Intel Corporation implements a strategy to address supply chain risks; Intel Corporation's Oregon wafer factory has started mass production of Intel 18A, and Fab 52 in Arizona has completed trial production, with plans to enter mass production ramp-up by the end of 2025. The 12nm node developed in collaboration with United Microelectronics Corp. Sponsored ADR (UMC) is expected to complete verification in 2026, with mass production planned for 2027, mainly serving the IoT and automotive electronics markets. In terms of capital investment, Intel Corporation has invested a total of $90 billion over the past four years, with $37 billion allocated to wafer factory equipment, resulting in a 100% increase in EUV capacity at the Arizona and Israel wafer plants. This aggressive capacity expansion strategy has increased automotive-grade storage encapsulation capacity by 50% in 2025, with monthly production capacity exceeding 15 million units. The Intel Foundry Accelerator Alliance has added two new projects: the Chiplet Alliance and the Value Chain Alliance. The Chiplet Alliance focuses on defining interoperable and secure chiplet standards to support customers in developing heterogeneous integrated products based on Chiplet technology, such as connecting Intel 18A-based computing dies with HBM high-bandwidth memory via a UCIe interface. The Value Chain Alliance integrates IP, EDA, and design service resources to provide customers with a one-stop solution from process design to mass production. Partners like Synopsys and Cadence have provided EDA tools and reference flows for Intel 18A. In terms of customer expansion, Intel Corporation's 16nm products in collaboration with MediaTek have entered wafer factory production, while deep collaboration in AI chip and autonomous driving domain controllers has been initiated with companies like Microsoft Corporation and Qualcomm. This ecological synergy model has increased Intel Corporation's automotive-grade storage revenue share to 12% in 2024, with plans to further increase to 15% in 2025. The current global wafer foundry market has a dominant pattern of "Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR leading, Samsung catching up, and Intel Corporation breaking through." According to TrendForce research data, Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR holds the top spot with a 67% market share, Samsung is second with 10%, and Intel Corporation only has 1%. In the field of AI chip production, Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR occupies 68% share, while Intel Corporation only has 5%. To break this pattern, Intel Corporation is adopting a strategy of "technological differentiation + ecological bundling." At the process level, Intel 18A reduces resistance loss through PowerVia technology, optimizing power consumption by 10% compared to Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR N2 node; in the packaging segment, Foveros Direct's 3D stacking technology achieves higher integration density, such as integrating CPU, GPU, and HBM memory in a single package, reducing the area by 40% compared to traditional 2D design. Currently, Intel Corporation's foundry strategy still faces three core challenges. First is the challenge of technical validation, where the yield improvement and customer certification progress of Intel 18A may not meet expectations; Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR N2 node has a yield rate of 80%, while Intel Corporation's 18A is still in the risk trial production stage. Second is the ecological barrier, where Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR locks in large customers like NVIDIA Corporation and AMD through a combination of "CoWoS + packaging + 3nm process", while Intel Corporation needs to break through trust barriers with customers in the AI chip field. Lastly, there is financial pressure, with Intel Corporation's foundry business experiencing a $13.4 billion loss in 2024. If the 18A mass production falls short of expectations, it could further worsen cash flow. The technological breakthroughs and ecological layout showcased by Intel Corporation at the 2025 foundry conference signify a strategic restructuring of the industry landscape through the dimensions of "process + packaging + manufacturing + ecology". Despite facing the monopoly pressure of Taiwan Semiconductor Manufacturing Co., Ltd. Sponsored ADR and internal financial challenges, Intel Corporation, through the mass production of 18A, 14A technology reserves, and global capacity expansion, is gradually establishing a differentiated advantage in the AI chip and automotive-grade storage fields. If it can continue to make breakthroughs in customer certification and yield improvement, Intel Corporation is poised to gain a larger market share and influence in the future.The corporation is expected to achieve the goal of becoming the "world's second largest contract manufacturing factory" by 2030, bringing new competitive factors to the semiconductor industry.This article is reprinted from Wallstreetcn, edited by GMTEight: Chen Yufeng.