Looking back at the history of industrial revolutions, the sources of national competitiveness have been repeatedly reorganized alongside changes in the core technologies and technological paradigms that define each era. The steam engine and railways transformed the spatial and economic order of the Industrial Revolution. Electricity and mass production created a new industrial society. Automobiles and oil formed the foundation of the twentieth-century industrial economy, while semiconductors, computers, the internet and mobile communications led to the rise of the information society. These technological revolutions did not merely grow one particular industry. They changed the methods of production across the economy, industrial structures, corporate organization, individual lifestyles and, ultimately, the very conditions of national competitiveness.
In economics, technologies that can be widely applied across many industries and that continuously stimulate technological innovation are called general-purpose technologies, or GPTs. The steam engine, electricity and information and communications technologies are representative examples, and today AI is also being evaluated as a new general-purpose technology. However, the meaning of the AI transformation cannot be fully explained by the concept of a general-purpose technology alone. AI is changing not only the way the economy produces goods and services, but also human intellectual labor, creation, the production and circulation of information, social decision-making and the nature of competition among states. In this sense, AI should be understood not only as a new general-purpose technology, but also as a foundational technology that reorganizes the competitiveness of existing industries and the productive capacity of nations.
Therefore, the economic value of AI cannot be assessed only by the size of the independent “AI industry.” What matters more is the multiplier effect of productivity improvement and innovation that occurs when AI is combined with existing industries. A new structure is emerging: AI competitiveness multiplied by existing industrial competitiveness equals new national competitiveness. For example, the future competitiveness of the automobile industry will not be determined by manufacturing capability alone. Autonomous driving, AI-based design and manufacturing, vehicle software and data capabilities will all operate together. In biotechnology, AI is being combined with drug development and medical data analysis. In the media and content industries, AI is entering the entire process of planning, production and distribution. In defense, AI is being integrated into intelligence analysis, command and control, unmanned systems and autonomous systems. In this sense, AI competitiveness is no longer merely the competitiveness of one independent industry. It is becoming a foundational capability that can amplify or weaken the competitiveness of a country’s existing industries.
AI competition is ecosystem competition.
This transformation is also changing the nature of AI competition among countries. In the early stage, the center of competition was which actor could develop better algorithms or AI models. Today, however, AI competition has evolved into a much more complex form of ecosystem competition. To develop powerful AI, advanced semiconductors and GPUs are required. To operate them at scale, data centers, cloud infrastructure and enormous amounts of electricity are needed. To train models, high-quality data is necessary, along with highly skilled talent capable of developing and applying them. Only when the resulting foundation models are connected to real services and innovation in industrial settings can they lead to productivity growth and economic expansion.
If a decisive bottleneck appears in any one part of this chain, a country’s AI competitiveness can be limited. Even if a country has world-class AI researchers, it will be difficult to develop frontier models without sufficient computing resources. Even if it develops excellent models, delayed adoption in industrial settings will make it difficult for those models to translate into economy-wide productivity growth. Conversely, even if a country’s own frontier-model capabilities are relatively weak, it can still generate significant economic effects if it rapidly adopts global AI technologies and combines them with the competitiveness of existing industries such as manufacturing, finance, healthcare and content.
For this reason, what matters in national AI strategy is not a simple ranking of technologies. National AI competitiveness depends on how effectively a country can secure and connect each element of the ecosystem, from advanced semiconductors and computing to energy, data, AI models, talent and industrial application. Ultimately, AI competition is moving beyond competition over individual technologies and becoming competition over the entire AI ecosystem.
AI competition is moving from technological competition to a contest for AI power.
What deserves particular attention in the way AI is reshaping national competitiveness is that AI competition no longer remains confined to technological development among companies or industrial competition. As AI development begins to affect not only economic growth and productivity, but also defense, intelligence, cybersecurity and scientific and technological capability, AI is rapidly becoming a strategic asset that shapes both economic power and security capability. As a result, competition over AI is expanding from technological competition to industrial competition, and then to national strategy and global power competition.
This shift is most clearly visible in recent U.S. AI strategy. The United States defines AI not merely as an advanced industry, but as a foundational technology that will determine future economic growth, national security and global competitiveness. In particular, the America’s AI Action Plan, announced in 2025, explicitly set “winning the AI race” as a national objective and presented three pillars of national strategy: accelerating AI innovation, building AI infrastructure in the United States, and securing leadership in international AI diplomacy and security.
In 2026, this trend is becoming even clearer. The United States is moving to rapidly introduce frontier AI models into national-security domains such as defense and intelligence, build next-generation high-performance computing infrastructure for national-security purposes, and connect private-sector advanced AI capabilities more directly with national-security systems. The effort to translate technological advantage in AI into economic competitiveness as well as military and strategic advantage is now taking shape as concrete national policy.
China is also moving quickly in response. China has positioned AI as a core driver of technological self-reliance, industrial upgrading and the transformation of the broader economy and society. Through its “AI Plus” strategy, China is seeking to diffuse AI across manufacturing, science and technology, consumption, healthcare, education and social governance, while building its own AI ecosystem that connects AI models, semiconductors, computing, data and industrial applications. Recent national development plans have also identified core AI technologies, advanced models and algorithms, high-performance semiconductors and cloud infrastructure, national data infrastructure and industry-specific AI applications as key national tasks.
At the same time, China is attempting to take a differentiated approach in global AI governance. By emphasizing openness and inclusiveness in AI, AI access for developing countries, cooperation with the Global South and UN-centered global AI governance, China is seeking to expand the international influence of its AI technologies and policy approach. Therefore, U.S.-China AI competition goes beyond a race over which country can develop the better AI model. The unit of competition is expanding from AI models to technology stacks, from technology stacks to industrial ecosystems and supply chains, and then to technical standards and global governance. Ultimately, AI competition is taking on the character of a new form of great-power competition in which technological advantage is converted into economic power, security capability and influence over the international order.
From this perspective, advanced semiconductors, computing, data centers, energy, data and AI models can no longer be viewed merely as ordinary technologies and commodities traded freely in international markets. Their character is changing into strategic assets that must be secured, protected and, in some cases, controlled from the perspective of national security and economic security. This is why technology export controls, outbound investment restrictions, supply-chain reorganization, data-transfer regulations, technical standards and international norms are all becoming instruments of AI power competition.
This also brings an important change to the assumption of the free movement of technology that has supported globalization and digitalization for decades. Security and sovereignty are now strongly intervening in a technology ecosystem that had been organized around efficiency and global specialization. It is precisely at this point that the question of AI sovereignty emerges. Of course, not every country can independently develop frontier AI models and semiconductors at the level of the United States or China. However, as economies, public administration, defense and social systems depend more deeply on AI, complete dependence on a particular country or a small number of global companies for core AI technologies and infrastructure can itself become a new strategic risk. Each country is therefore facing a new policy choice: while participating in the global AI ecosystem, what level of technology, computing power, data, language capability and cultural autonomy should it secure for itself?
In this way, AI has moved beyond being a single advanced technology and has become a core foundation of national competitiveness that simultaneously shapes economic power, security capability and strategic position in the international order. Accordingly, national responses to AI must also expand beyond support for technology development, industrial promotion and risk regulation. They must encompass the economy, society, security and diplomacy.
As AI becomes a question of national competitiveness, national security and global power, AI policy must be redefined as a comprehensive national strategy that determines a country’s future competitiveness and strategic position.
