This post is also available in:

AI Has Already Begun, But the Starting Line is Not Equal

Artificial Intelligence (AI) is no longer a future technology. It has already moved deep into people’s daily lives, labor markets, education, and research sites, establishing itself as a key variable in competition between nations. The recent spread of technology centered on Generative AI is further accelerating these changes. The problem is that this change is not given to all countries under the same conditions. The World Bank’s 2025 publication, Digital Progress and Trends Report 2025: Strengthening AI Foundations, clearly reveals that the world facing the AI era is by no means flat. The core diagnosis of this report is that while AI appears to be a universal technology, it is actually spreading on extremely uneven foundations, and this gap is becoming more structural over time. According to the report, high-income countries account for the vast majority of AI models, startups, investment capital, and research achievements worldwide. As of 2025, more than 80% of notable AI models were developed in high-income countries, and AI startups and venture capital are also concentrated in a handful of nations. Conversely, while low- and middle-income countries are rapidly being incorporated as consumers and users of AI technology, they have almost no access to positions where they can design the technology or determine its direction. On the surface, the global spread of tools like ChatGPT seems to show the democratization of AI, but behind it, the old imbalance of “who creates and who uses” is being reproduced in a new form.

The Question Posed by the World Bank: “What Sustains the Foundation of AI?”

The point where this report distinguishes itself from existing AI discourse is that it redefines AI competition not as a matter of technological level or algorithm performance, but as a matter of “fundamental strength.” The World Bank summarizes the conditions that enable the spread of AI into four elements and names them the “4Cs.” These refer to Connectivity, Compute, Context, and Competency. These are significant because they are not problems that can be solved in the short term with individual technologies or single policies; rather, they reflect a nation’s digital, educational, and institutional foundations accumulated over a long period.

The first element, Connectivity, goes beyond simple internet penetration to include stable power supply, high-speed networks, and access to digital devices. The second, Compute, refers to access to computing resources—the “new electricity” of the AI era—such as semiconductors, data centers, and cloud infrastructure. The third, Context, refers to the ability to utilize AI while reflecting local conditions such as data, language, institutions, and social trust. Finally, Competency indicates the level of human resources capable of understanding, utilizing, and developing AI. The World Bank diagnoses that if any one of these four is weak, it is difficult for AI to spread throughout society, regardless of its technological potential.

Notably, this “fundamental strength” is something that cannot be caught up within a short period. While AI technology itself can be quickly replicated and transferred, the infrastructure, education systems, and talent cultivation structures that support it are the results of long-term policy choices and accumulated investment. Because of this, AI competition is proceeding with the starting lines already far apart, and latecomers face a “strength gap” before they even face a “technology gap.” The World Bank warns that at this very point, inequality in the AI era may solidify into a structural gap rather than a simple technological one.

The Option of “Small AI”: A Leapfrog or a Stopgap?

One of the important concepts presented in the World Bank report is “Small AI.” This refers to the use of relatively lightweight AI designed to operate even in everyday devices and limited computing environments, rather than ultra-large language models or cutting-edge AI requiring massive computational resources. AI focused on solving specific problems, such as crop management in agriculture, primary education assistance, and improving local health services, falls into this category. The World Bank evaluates “Small AI” as a realistic path that can provide immediate and practical effects for low- and middle-income countries. However, the report simultaneously points out the limitations of this strategy. While Small AI can alleviate accessibility and cost issues in the short term, it is difficult for it to become a fundamental alternative for raising a nation’s AI capabilities. This is because the core models, technological standards, and data ecosystems remain dominated by high-income countries and large corporations. Ultimately, Small AI could be a “stepping stone for a leap,” but if chosen without sufficient strengthening of fundamental strength, it carries the risk of entrenching technological dependence. This is why the World Bank treats Small AI cautiously as “one of the strategic options” rather than a complete solution.

광고

Why the AI Gap Culminates in Education and Competency

A point the report repeatedly emphasizes is that the AI gap eventually converges on the issue of talent and education. Utilizing AI does not end with simply knowing how to use a tool. To apply AI to organizations and industries, combine it with existing work, and develop it into new problem-solving methods, a certain level of digital literacy and analytical ability is required. The World Bank explains this as the “layering of AI competency,” pointing out that everything from basic AI literacy to mid-level application competency and high-level research and development competency is necessary. The problem is that the distribution of this competency is extremely uneven across countries. In low-income countries, even the proportion of the population with basic digital skills is very low, and talent with mid-to-high-level AI competency is extremely scarce. In such situations, even if AI tools are introduced, the human foundation to interpret, improve, and expand them is inevitably weak. Ultimately, AI remains a means of passively consuming technology given from the outside rather than a tool for increasing productivity. The World Bank warns that approaches attempting to solve the AI gap through simple technology transfer or equipment support are highly likely to fail. This analysis naturally brings the discussion to education systems, particularly higher education and vocational training. Competitiveness in the AI era is not determined by the achievements of a few high-tech research labs or startups. The long-term capability of a nation depends on how much of the population can learn in a digital environment, understand AI, and continuously acquire changing technologies. The reason the World Bank emphasizes “Competency” as the core of AI policy is that AI competition is ultimately a matter of institutions and education surrounding people.

Universities Emerging as the Bottleneck of AI Fundamental Strength

In the World Bank report, “Competency” does not mean simple technical proficiency. It is a concept that includes the ability to understand and utilize AI, the learning ability to catch up with and relearn technological changes, and the organizational experience of applying AI to new problems. Based on this definition, universities are bound to become core strongholds of AI fundamental strength. This is because higher education is a space where talent forms thinking patterns and expertise over a long period, beyond short-term technical training. Nevertheless, the report points out that in many countries, universities are failing to perform this role properly. Especially in universities in low- and middle-income countries, cases where digital and AI-related curricula fail to keep pace with industrial changes are widespread. In some regions, many universities have not revamped their curricula for more than five years, and practical infrastructure for AI, data analysis, and computing is limited. This reveals a structural problem that the entire higher education system is not designed to respond to technological changes, rather than a simple matter of budget shortage. Consequently, universities face the risk of degenerating into institutional spaces increasingly alienated from the capabilities required by the market, rather than spaces for nurturing talent for the AI era.

One of the most uncomfortable messages the report throws out is the fact that the center of AI innovation is already leaving the university. Cutting-edge AI research and model development have increasingly required large-scale compute resources and vast data accessibility, and the entities capable of meeting these conditions are primarily global big tech companies and a few research-intensive institutions. Many universities are unable to participate fully in this competition due to financial and infrastructure constraints, and as a result, the leadership in AI research is shifting from the academic community to industry. This change does not end as a problem of the research environment. As universities are pushed out of the center of AI research and education, the channels for systematically cultivating advanced AI competency also narrow. This leads to a vicious cycle of talent shortage and technological dependence. At this point, the World Bank redefines the problem surrounding AI as a matter of “the ability of institutions and organizations to absorb change,” rather than the “speed of technology adoption.” If universities fail to change, competition in the AI era will naturally be determined outside of them, and higher education will inevitably be reduced to a role of catching up with the results too late.

AI Fundamental Strength is Not Built Overnight

The message the World Bank report repeatedly emphasizes is clear. The AI gap is not a problem that can be solved by technology dissemination or short-term projects. AI fundamental strength—composed of Connectivity, Compute, Context, and Competency—is the result of policy choices, investments, and education systems accumulated over a long period. Therefore, AI competition is closer to a matter of direction and structure than a matter of speed. A nation’s position is determined by what systems it has maintained and strengthened, rather than what it introduces right now. This is why the report defines AI as a “new general-purpose technology” while distinguishing it from previous technological innovations.

Particularly, the realms of competency and education are the most decisive factors, while their effects appear the slowest. Nurturing talent, transforming education systems, and making universities and research institutes adapt to technological changes require years, sometimes decades. Countries lagging in the AI era are likely the result of missing opportunities to accumulate fundamental strength for a long time, rather than simply introducing technology late. What the World Bank warns through this report is precisely this “time gap.” It means that AI competition is not in the beginning stage, but has already entered a phase where the difference in accumulation is becoming apparent.

In this context, the core question of the AI era naturally converges on universities and the higher education system. The university is a space that can systematically accumulate “Competency” among the AI fundamental strengths, while also being the institution that has reacted most slowly to change. The reality that AI does not wait for the university means that higher education can no longer remain on the periphery of technological change. If universities cannot redefine themselves as the center of the AI era, a nation’s AI strategy is highly likely to remain an empty slogan.

The World Bank’s diagnosis examined in this first installment is not a simple international comparison or technological outlook. it is a structural question asking what choices nations and universities have made—and what choices they have postponed—in the AI era. If the essence of AI competition is fundamental strength rather than models or algorithms, the answer to where and how that strength should be cultivated also becomes clear. The next installment will put this question directly to the university. In the AI era, is the university still at the center, or has it already been pushed to the periphery?

#AI #AIFundamentalStrength #WorldBankAIReport #DigitalProgressAndTrendsReport2025 #AIDivide #AIInequality #4CsFramework #NationalAIStrategy #AICompetency #DigitalInfrastructure #HigherEducationAndAI #UniversityAICrisis #AITalentDevelopment #AIPolicyAnalysis #SpotlightU