BOK governor opens regional economy symposium

Complementary innovation in infrastructure, institutions needed

Physical AI may ease concentration in Greater Seoul

Bank of Korea Governor Shin Hyun-song speaks at a Monetary Policy Board briefing on monetary policy direction at the Bank of Korea in Jung-gu, Seoul, on Aug. 27. [Yonhap]
Bank of Korea Governor Shin Hyun-song speaks at a Monetary Policy Board briefing on monetary policy direction at the Bank of Korea in Jung-gu, Seoul, on Aug. 27. [Yonhap]

Bank of Korea Governor Shin Hyun-song said Tuesday that AI holds the potential to fundamentally transform the foundations of productivity over the long term, just as the steam engine, electricity and the internet once did.

Shin made the remarks in an opening address at the BOK Regional Economy Symposium held Tuesday afternoon at Lotte City Hotel in Daejeon, where he described AI as a general-purpose technology rapidly spreading beyond specific industries to permeate the entire economy.

Citing the dynamo — the first industrial electric generator — Shin said that when factories switched from steam to electric power in the late 19th century, people expected a swift productivity surge, but the gains were slow to materialize because factory layouts were still designed around steam-era systems. "The productivity benefits only emerged in full once factories were fundamentally redesigned around electricity," he said.

Shin invoked economist Paul David's argument that the economic effects of a new general-purpose technology appear not when the technology is adopted, but when production systems are reorganized around it.

"Today's AI is no different," he said, stressing that translating AI adoption into productivity gains requires "complementary innovation" in technology infrastructure, organizational structures and institutional frameworks. Beyond installing new software and building data centers, he said, workers must develop new capabilities, companies must change how they operate, and AI infrastructure must be linked to local industries and markets.

"Ultimately, whether AI becomes a new growth engine for regional economies depends less on how quickly the technology is adopted and more on how regional economies are redesigned to maximize productivity through AI," Shin said.

The Regional Economy Symposium has been held since 2023 as a forum for business, academic and government representatives to discuss regional economic issues and explore policy responses.

At the same event, Jeon Min-su, head of the BOK's regional economy research team, presented findings on AI's potential to widen labor market disparities across regions, under the theme "AI and Regional Labor Markets: Risks of Widening Interregional Gaps and New Opportunities."

The BOK's analysis of regional AI exposure — a measure of how readily AI can be applied to local jobs — found that Seoul ranked highest for generative AI, followed by Sejong, Gyeonggi Province and Incheon. The pattern was similar for agentic AI, with Seoul leading, followed by Sejong, Gyeonggi Province and Daejeon. Physical AI exposure, by contrast, was highest outside the greater metropolitan area, led by North Gyeongsang Province, South Jeolla Province, North Chungcheong Province and Ulsan.

Jeon attributed the pattern to the concentration of knowledge-service industries in the greater metropolitan area, which supports large numbers of office, sales, managerial and professional jobs with high exposure to generative and agentic AI. Non-metropolitan regions, where manufacturing and agriculture, forestry and fisheries account for a larger share of the economy, have more equipment-operation, manual labor and service jobs with high physical AI exposure, he said.

Jeon then divided the regional labor market implications of AI's spread into two categories: risks and opportunities.

On the risk side, he said AI development could reinforce the agglomeration economies of knowledge services and semiconductor manufacturing already concentrated in the greater metropolitan area. He also warned that if physical AI advances accelerate, employment in non-metropolitan regions could decline, particularly in manufacturing, accommodation, food service and transportation.

On the opportunity side, Jeon said advances in human-AI collaboration models and physical AI could ease the concentration of talent and research and development in the greater metropolitan area. AI could also deliver larger productivity gains in non-metropolitan service industries that face human capital shortages, easing supply constraints and spurring growth in local service consumption, he said. He added that if AI's spread generates relatively high-quality jobs in non-metropolitan areas, it could slow the outflow of young people from those regions.

Jeon recommended that policymakers actively pursue stronger AI education systems to promote service-sector specialization in regional economies, support for innovation-driven growth, development of service specializations linked to key manufacturing industries to help regional manufacturers shift toward knowledge services, cultivation of startups, and focused investment to transform regional hub universities into AI centers.


kimstar@heraldcorp.com