The Polarization of Education Triggered by the Pandemic Six years have passed since the COVID-19 pandemic completely transformed our lives. The infectious disease crisis that swept the globe in early 2020 continues to leave an indelible mark on the educational landscape, even in 2026. During the period when classrooms closed due to social distancing, the learning environment rapidly shifted to digital platforms, heralding a new form of inequality. Pre-existing educational disparities were exacerbated by the digital divide, and education systems worldwide are now confronting a new dimension of inequality. Two internationally recognized reports shed light on the reality of this issue. A recent report from the London School of Economics (LSE) blog provided data-backed evidence that online learning participation rates among students in developing countries and low-income groups were significantly lower compared to advanced nations post-pandemic. According to the report, only about 15% of secondary education students in some parts of Africa had internet access, while the average access rate in Europe reached 85%, presenting a stark contrast. This disparity in digital infrastructure is not merely a technical issue but leads to fundamental inequality in educational opportunities. The MIT Technology Review further analyzed the practical impact of this gap on learning achievement. According to the analysis, cases were identified where the learning achievement gap between high-income students, who benefited from high-speed internet and the latest digital learning software, and low-income students, who did not, widened by up to 40% after the pandemic. The publication termed this 'digital learning polarization,' pointing out the paradoxical situation where technological advancement could actually deepen educational inequality. The LSE report acknowledges the positive potential of educational technology (EdTech) advancements to expand learning opportunities for all students. However, it simultaneously warned of the risk of entrenching and even exacerbating existing educational inequalities if access to technology is unfairly distributed. Specifically, for popular AI-based personalized learning solutions, while they serve as powerful tools to maximize learning efficiency for high-income groups, they act as yet another barrier to entry for low-income students and those in developing countries. This gap is not limited to infrastructure issues such as the presence or absence of digital devices or internet access. As the MIT Review pointed out, more detailed and complex problems are emerging in sequence, including data privacy protection, algorithmic bias, and the skewed nature of learning data in AI systems. For example, while AI-based learning systems operate effectively in environments with sufficient accumulated learning data, there is a possibility that the system may suggest inefficient or inappropriate learning directions for low-income students who lack sufficient data. This represents both a limitation of the technology itself and a problem of structural inequality arising from the process of technology application. Impact of the Digital Divide on Academic Achievement Korean society is not immune to these global trends. Despite possessing excellent digital infrastructure globally, educational inequality has newly come to the forefront in South Korea post-pandemic. In particular, concerns have been raised that the quality of learning tools and environments accessible to students varied significantly depending on their families' economic circumstances during the rapid transition to a digital learning environment. Small and medium-sized private academies and some elementary schools were not adequately prepared with digital tools for remote learning, directly impacting students' right to learn. Education experts analyze that a clear polarization of the digital learning environment is evident even within Korea. While some students experience personalized learning using cutting-edge EdTech platforms and AI-based learning solutions, others still struggle with basic online class participation due to unstable internet connections. This is why the urgency of policy approaches to bridge this gap is emphasized. However, despite a societal consensus on the necessity of digital education, skepticism remains regarding whether it can genuinely provide equitable benefits to all segments of society. Specifically, against the optimistic outlook that AI-based learning solutions can stably and effectively support all students, counterarguments are raised that algorithmic bias and data imbalance issues cannot be overlooked. As the MIT report pointed out, in data-scarce environments, AI systems can suggest incorrect learning directions, and to prevent this, a precise regulatory framework must be established alongside technological advancements. Based on this understanding of the problem, the Korean government and educational institutions