Government Launches AI Mega-Project… “Expanding Beyond Semiconductors to Power and Robotics”
Yuanta Securities Korea Report
Development of a Second Semiconductor Hub in the SuNAM Co., Ltd. Region and Promotion of HBM Back-End Processes in the Chungcheong Region
Infrastructure Expansion Aims for 18.4 GW of AI Data Center Capacity by 2035
Yuanta Securities: “Phase 2 of AI Capex… Benefits the Entire K-Manufacturing Sector”
[Edaily Reporter Park Sun-Yeop ] As the government announced the “Three Major Mega-Projects for Korea’s Great Leap Forward,” centered on semiconductors, physical artificial intelligence (AI), and AI data centers, expectations are growing in the stock market that the AI investment cycle will expand. While large-cap semiconductor stocks have led the AI rally so far, driven primarily by rising prices for High-Bandwidth Memory (HBM) and memory chips, analysts suggest that the investment theme could broaden to include data centers, power grids, power generation and transmission, cooling systems, robotics, back-end processes, and materials, components, and equipment. In a report released on the 30th, Lee Jae-won, an analyst at Yuanta Securities Korea, interpreted this policy as “Phase 2 of AI Capex.” He explained that while Phase 1 was a market driven by large-cap semiconductor stocks due to rising HBM and memory prices, Phase 2 marks a phase where benefits spread across the entire industry as physical bottlenecks required for AI infrastructure construction are resolved. (Chart: Yuanta Securities Korea)
The three major megaprojects proposed by the government consist of semiconductors, physical AI, and AI data centers. In the semiconductor sector, the government plans to develop the Southwest region as a second semiconductor production hub following the Seoul Metropolitan Area and push for the early completion of the existing clusters in Yongin and Pyeongtaek. The Chungcheong region will be developed as a hub for HBM back-end processes. Yuanta Securities Korea believes this will not only help maintain a significant lead in memory technology, centered on SamsungElectronics and SK hynix, but also broaden the scope of beneficiaries to include HBM, back-end processes, materials, parts, and components, and power semiconductors. The scale of investment is also substantial. According to the report, based on government announcements, the investment for the four semiconductor fabs in the Southwest region totals 800 trillion won, while HBM packaging in the Chungcheong region amounts to 81 trillion won. Additionally, 550 trillion won is earmarked for AI data centers, and 30 trillion won for next-generation semiconductor research and development (R&D). SamsungElectronics(005930)Including existing investment plans, the report mentioned investment plans totaling 2,030 trillion won for Pyeongtaek and Yongin, 425 trillion won for the Honam region, 140 trillion won for the Chungcheong region, and 60 trillion won for the Yeongnam region. SK hynix(000660)presented figures of 1,000 trillion won for AI data centers (AIDC), 600 trillion won for DRAM in Yongin, 100 trillion won for NAND in Cheongju, and 400 trillion won for the SuNAM Co., Ltd. region. The Physical AI Project focuses on the AI transformation of manufacturing and the commercialization of AI robots. The government plans to establish data factories, develop its own Physical AI foundation model, and localize the production of robot components. The plan also includes goals to deploy more than 1,000 AI robots annually, train 10,000 AI robot specialists, and develop an independent model within three years. The researcher assessed this as a trend in which AI is expanding beyond the realms of software and the cloud into real-world industries such as manufacturing, logistics, caregiving, agriculture, safety, and national defense. AI data centers are also a key pillar. Efforts are underway to build large-scale AI computing infrastructure centered around SK(034730), GS Holdings(078930), and #Naver, and plans were presented to integrate these with domestically produced neural processing units (NPUs), power and cooling solutions, and cloud technology. The target capacity is 18.4 GW by 2035. The researcher noted that as data centers grow, bottlenecks shift from computational capacity to the power grid, power generation, storage, and cooling, suggesting that power equipment, power generation facilities, energy storage systems (ESS), and data center equipment and solutions could be key beneficiaries. Yuanta Securities Korea viewed the key investment appeal of the Korean stock market as lying not merely in Korea’s status as an AI-using nation, but as a supplier nation that channels AI investment demand into manufacturing profits. The firm explained that Korea possesses a rare manufacturing supply chain that simultaneously encompasses memory semiconductors, power equipment, nuclear power and power generation facilities, and rechargeable batteries. It also cited the shift of parts of the China-centric supply chain—particularly in high-value-added and high-reliability sectors—toward Korea following the U.S.-China conflict as a structural benefit. However, there are still significant hurdles to overcome before the project can be fully implemented. Securing power, water, land, and labor could act as bottlenecks, and in the case of semiconductors, concerns about long-term oversupply may arise. The analyst predicted, “It is unlikely that this project will have an immediate impact on resolving the short-term memory supply shortage,” adding, “Short-term stock price reactions will vary by sector depending on the actual pace of construction, the securing of power infrastructure, board approvals at individual companies, and demand visibility, rather than the total investment amount.”
With SamsungElectronics’ “Together with the People: SamsungElectronics Appreciation Festival,” which began on the 8th of last month, coming to a close on the 5th, there are expectations that this even…
Hyundai Department Store ( HYUNDAIDEPARTMENTSTORECO.,LTD(069960)) announced on the 5th that it received the highest rating of “AA” in the first half of this year’s ESG evaluation conducted by SustainB…
The way global companies adopt artificial intelligence (AI) is rapidly shifting toward a field-based engineering approach. As the ability to successfully integrate AI into actual business systems beco…