National strategy for the AI revolution beyond the threat of new Luddites

Governments should avoid delaying productivity gains from AI adoption. Expanding the economic pie quickly offers the best chance to cushion distributional losses.

Kwon Nam-hoon

The author is the president of the Korea Institute for Industrial Economics and Trade.

The global surge in AI has intensified. Just two years ago debate centered on whether AI development should be restrained. Today, the world is in the midst of massive infrastructure investment and strategic competition for technological leadership.

Five hyperscale technology firms — Alphabet, Amazon, Meta, Microsoft and Oracle — plan to spend a combined $560 billion in capital expenditures this year alone. The scale is comparable to the GDP of a midsize economy.

Boston Dynamics Atlas robots are displayed in the Hyundai Motor Group booth during CES 2026 in Las Vegas, Nevada on Jan. 6. [REUTERS/YONHAP]

AI itself is evolving rapidly. Beyond generative systems that write and draw, “agent AI” can now make decisions and perform sequences of tasks autonomously. More recently, the technology has expanded into “physical AI” deployed in manufacturing and robotics. Humanoid robots unveiled at this year’s CES event triggered labor union concerns, with some observers warning of a new wave of worker resistance reminiscent of the 19th century Luddite movement.

Whether AI and robots will replace human labor entirely remains uncertain. Yet waiting for change to fully unfold before preparing would be risky. The trajectory of technological progress suggests the need for early and deliberate adaptation.

llustration depicting the 19th-century Luddite movement in Britain. [WIKIPEDIA]

Technology has always evolved alongside the way humans create value. Economist J. Bradford DeLong classified human contribution into six domains: physical strength, manual precision, communication, calculation and control, creativity and emotional connection. Each wave of innovation has replaced or enhanced some of these functions.

Machines took over physical labor and repetitive manual work. Media and communications amplified the human voice. Computers extended cognitive processing. Creativity and empathy were long considered uniquely human, but AI is beginning to challenge even these boundaries through advanced reasoning and increasingly natural interaction.

Still, AI will likely follow the familiar dual path of substitution and complementarity. A joint study by the International Monetary Fund and the Bank of Korea estimates that about half of Korean workers will be directly affected. Of them, roughly half may face displacement pressure, while the rest could see higher productivity and income by working alongside AI.

Which jobs disappear and which survive depends less on occupations than on tasks. AI is rapidly absorbing structured work such as basic translation, entry-level coding, data organization and contract review. Jobs composed largely of such tasks face higher risk. Where work includes complex judgment, coordination or client interaction, AI can instead become a powerful tool.

History offers a useful example. When automated teller machines were introduced, many expected bank teller jobs to vanish. Economist James Bessen found that overall employment did not decline. Routine cash handling fell, but tellers shifted toward customer service, loan consulting and financial guidance. Task composition changed, not employment itself.

Concerns have also emerged that AI could eliminate entry-level work and weaken the “experience ladder” for young workers. A recent Stanford University study found a 16 percent decline in employment among U.S. college graduates in their twenties in AI-exposed occupations. Yet early research also suggests that AI can narrow performance gaps by helping less experienced workers improve their output.

More fundamentally, some worry that AI could shift the balance between labor and capital. Philosopher Michael Polanyi described “tacit knowledge” as the skills people know but cannot fully explain. Traditional automation struggled to capture this. AI, however, can learn from data and observation, potentially transferring workers’ skill into corporate systems. Such a shift could reinforce the concentration of income and wealth long associated with capital-intensive technologies.

The original Luddites were not simply opponents of machinery. Skilled artisans protested technologies that enabled lower-skilled workers to replace them, threatening the value of accumulated expertise. Their experience illustrates a broader lesson: social systems must evolve as quickly as technology.

AI is a general-purpose technology capable of reshaping entire industries. Infrastructure takes time to build, but once deployed, its economic impact can be swift. The current investment surge may represent a narrow window for preparation.

Several policy directions stand out. Governments should avoid delaying productivity gains from AI adoption. Expanding the economic pie quickly offers the best chance to cushion distributional losses.

Regulatory reform is also essential. Barriers that limit new industries and markets should be eased, while frameworks such as data property rights must be clarified to support innovation and fair competition.

The Tesla humanoid robot Optimus is displayed at the Viva Technology conference dedicated to innovation and startups at Porte de Versailles exhibition center in Paris, France, June 12, 2025. [REUTERS/YONHAP]

Labor markets need a balance between flexibility and security. Work force mobility should be improved, but displaced workers require retraining and stronger social safety nets. Education must also shift away from rote knowledge toward critical thinking, emotional skills and lifelong learning that machines cannot easily replicate.

Tax systems should be adjusted in advance as income shifts toward capital. Incentives for corporate AI investment can coexist with reforms that strengthen asset formation and income stability for households.

Time for adjustment is limited. If policy remains constrained by outdated regulation or stakeholder conflict, societies may face the full force of disruption without preparation. A coordinated and forward-looking national strategy is essential to navigate the AI transition.

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.