Kim Yong-beom, Presidential Office Policy Chief, stated on the 10th that AI-era semiconductor competition has shifted from technology development to production capacity competition, emphasizing that the most critical resource governments must provide to companies is 'time' rather than fiscal support. In a Facebook post titled 'Production Capacity is the New National Power,' Kim argued that national competitiveness now depends more on how quickly and at what scale technology can be converted into production capacity than on who develops technology first. This shift is driven by structural increases in high-performance memory demand from AI data centers, autonomous vehicles, and humanoid robots, making production capacity expansion a strategic necessity in the AI revolution.
Kim Yong-beom Identifies Production Capacity as Core National Competitiveness Factor
Kim wrote that "the AI revolution is not simply a technological innovation but another production revolution that amplifies human knowledge and judgment." He stated that "national competitiveness is now determined more by how quickly and at what scale technology can be converted into production capacity than by who develops the technology first." Kim explained that AI competition leads to semiconductor competition, which in turn results in production capacity competition, due to structural increases in high-performance memory demand from AI data centers, autonomous vehicles, humanoid robots, and various AI devices.
![Kim Yong-beom speaking at a forum]()
Kim Yong-beom, Presidential Office Policy Chief, speaking at a forum
Traditional Chicken Game Strategy May Lose Effectiveness in AI Era
Kim diagnosed that the so-called 'chicken game' strategy that worked in the past memory industry may no longer be effective in the AI era. He stated that "in the past, a strategy where leading companies lowered prices through large-scale facility investments and volume offensives to push out latecomers from the market was effective," but added that "in an environment where memory demand is structurally increasing due to the AI revolution, it is difficult to conclude that the same strategy will have the same effect as before."
Kim explained that "if domestic memory companies significantly increase production, but demand from AI data centers, autonomous driving, robots, and various AI devices absorbs a significant portion of this, prices may not fall to levels that push out latecomers as in the past." He also analyzed that "some latecomers have conditions to endure longer investment periods than general market logic based on national support and domestic markets," noting that "the effectiveness of strategies to stop or eliminate latecomer investments through simple price competition alone may be more limited than in the past."
Strategic Shift from Price Competition to Eliminating Supply Gaps
Kim emphasized that "the strategy needed now is not to suppress competitors with price competition after they grow, but to prevent the creation of supply gaps that would allow competitors to grow in the first place." He stated that "when the overall market is growing rapidly, if the speed of production capacity increase cannot keep up with the market growth rate, market share may actually fall even if absolute production increases."
Kim explained that "fab expansion is not simply a growth investment but a strategic investment to connect technological superiority to market share and prevent latecomers from securing customers and volume." He stressed that securing production capacity directly translates to national competitiveness.
Government Role Focuses on Removing Infrastructure Bottlenecks
Kim repeatedly emphasized the importance of 'time' regarding the government's role. He stated that "in the AI-era production capacity competition, the more important resource that only the state can supply is time," pointing out that "companies can build fabs and invest in production facilities, but cannot solve power grids, water supply, transmission networks, national industrial complexes, and complex permit procedures on their own."
Kim stated that "the role of the state is not to invest on behalf of companies, but to timely remove bottlenecks that companies cannot solve themselves so that companies can continue long-term strategic investments," adding that "time is created only when actual bottlenecks such as power grids, water supply, transmission networks, and permits are actually removed."
He explained that "a three-year delay in securing production capacity does not just mean a delay in factory completion," noting that "during that period, customers and markets are handed over to competitors, and competitors use the secured sales and cash flow to expand R&D and facility investments again." Kim added that "while time can be compensated for with cost, lost markets cannot be easily recovered," concluding that "technology is not completed by invention but by production. National power in the AI era starts with technology but is completed with production capacity."
FAQ
What did Kim Yong-beom say about government support for semiconductor companies on the 10th?
Kim Yong-beom, Presidential Office Policy Chief, stated on the 10th that the most critical resource governments must provide to semiconductor companies is 'time' rather than fiscal support. He emphasized that the government's role is to remove bottlenecks such as power grids, water supply, transmission networks, and complex permit procedures that companies cannot solve on their own, enabling companies to continue long-term strategic investments in production capacity expansion.
Why does Kim Yong-beom believe traditional chicken game strategies may not work in the AI era?
Kim explained that in an environment where memory demand is structurally increasing due to AI data centers, autonomous vehicles, humanoid robots, and AI devices, increased production by leading companies may not lower prices to levels that push out latecomers as in the past. He noted that some latecomers have conditions to endure longer investment periods based on national support and domestic markets, making simple price competition strategies less effective than before.
