Korea's AI strategy overweights chips and data centers and believes that productivity, talent and manufacturing-based AI will decide the country's future.
Alice M1, a humanoid robot developed by Korean robotics startup A-Robot, demonstrates a vision recognition-based AI learning process at NextRise 2026, a startup fair at Coex in Gangnam District, southern Seoul, on June 18.NEWS1
Kim Byung-yeon
The author is a chaired professor of economics at Seoul National University.
In late June, the Lee Jae Myung administration announced three “megaprojects” involving some 1.5 quadrillion won ($1.02 trillion) in new investments in semiconductors, physical AI and AI data centers. Including existing investment plans by major corporations, the total reaches 5 quadrillion won. The government calls these sectors the “three pillars” of the AI era and hopes they will make Korea “irreplaceable.”
I share the government’s determination to respond to the historic transformation AI will bring. But these pillars are badly unbalanced and fail to identify where the decisive competition lies. More than 900 trillion won of new investment is earmarked for semiconductors and 550 trillion won for AI data centers, while public and private investment in physical AI totals only around 20 trillion won. Given physical AI’s enormous potential, this strategy seems trapped in the past.
Semiconductors alone cannot reverse the decline in Korea’s potential growth rate. Long-term improvements in living standards come from productivity — the ability to create more value with the same capital and labor. Yet Korea’s productivity is only about two-thirds that of Germany’s.
Why is Korea, home to the world’s leading memory-chip factories, so much less efficient than another manufacturing powerhouse? Windfall gains from soaring memory prices are different from productivity improvements across the economy. As modern growth theory emphasizes, innovation comes not from a boom in one component or industry but from technological development and general-purpose knowledge that spread throughout industries. Without them, Korea remains an input-driven economy that survives only by saving more and working longer.
Semiconductor investment must continue. Amid geopolitical upheaval, chips are not merely exports but critical economic-security assets. Advanced memory is a choke point in the global semiconductor supply chain. During upturns, it generates enormous cash flow, tax revenue and investment capacity. The memory cycle in the AI era will also differ significantly from previous cycles.
But designing an entire national growth strategy around memory leaves the country dangerously dependent on one industry and a narrow set of technologies. Betting the nation on particular products and industries, when technological paradigms are shifting rapidly and China is catching up, exposes the entire economy to severe volatility.
Physical AI is the key to raising Korea’s potential growth rate. Among advanced democracies, Korea is unusual in possessing a broad, sophisticated manufacturing ecosystem spanning semiconductors, automobiles, batteries, shipbuilding, steel, electronics and defense.
Manufacturing-based AI combines robots, sensors, process data, skilled engineers and real production sites to train machines and continually improve performance. Korea is exceptionally well positioned for such dynamic learning. If these technologies spread beyond individual industries into other manufacturing and service sectors, the productivity gains could far exceed temporary booms enjoyed by particular products or conglomerates. Physical AI offers a rare opportunity to transform the economy’s underlying structure.
Manufacturing AI would also increase Korea’s strategic value. A country that goes beyond supplying memory chips to organically integrate semiconductors with physical industries through AI gains far greater geopolitical leverage. If Korea becomes a manufacturing hub built around physical AI, its economic and security value within the democratic bloc could become unparalleled. Korea could become a pivotal partner supplying AI robots to factories in the United States and other allied countries.
The alternative is troubling. If Korea falls behind while China combines AI and robotics with its vast manufacturing base, then dominates technological standards and supply chains, Korean manufacturers could eventually become dependent on Chinese AI platforms. If China becomes the unrivaled winner in physical AI, the global economy and ultimately the geopolitical balance could rapidly tilt toward Beijing.
At the pinnacle of this strategy must be attracting talent. Ultimately, people generate new ideas and develop general-purpose technologies that produce dramatic productivity gains. With universities and research institutes in the United States and elsewhere cutting recruitment and research positions under financial pressure, Korea has an opportunity. Korea is among the few advanced economies currently enjoying substantial growth in tax revenue.
By reprioritizing national resources, dramatically improving research conditions at universities and institutes and recruiting postdoctoral researchers and established scholars worldwide with stable employment for three to five years or longer, Korea could make one of the most cost-effective investments available for raising potential growth.
Korea’s growth has long come from building the world’s best factories. But if it simply repeats that strategy, it may eventually become not the creator in “creative destruction” but the one being destroyed.
The government must now use physical AI to raise productivity throughout industry, cultivate, retain and attract talent and ensure that new ideas are born in Korea. That is where the real contest for Korea’s economic future will be decided.
This article was originally written in Korean and translated by a bilingual reporter with the help of generative AI tools. It was then edited by a native English-speaking editor. All AI-assisted translations are reviewed and refined by our newsroom.
