인공지능 활용 확대에 따른 법치주의 실현의 과제 = Challenges of Realizing the Rule of Law Amid the Expanding Use of Artificial Intelligence

다국어 초록 (Multilingual Abstract)

Artificial intelligence (AI) is already being utilized across various aspects of daily life and will continue to drive innovation. However, no degree of innovation can supersede the rule of law. As the exercise of public authority through AI expands, a thorough examination and appropriate response to its impact on the rule of law are required. This paper analyzes the potential effects of AI-powered public authority on the rule of law and explores normative measures to ensure its realization.
The rule of law is fundamentally based on a distrust of human discretion. At its core, it mandates that administrative actions be grounded in formal legislation enacted by the legislature. It seeks to ensure the accuracy, rationality, fairness, objectivity, neutrality, and consistency of governmental actions by requiring adherence to specific forms, procedures, and standards. Furthermore, it establishes mechanisms for evaluating the legitimacy of government actions and holding authorities accountable. AI, characterized by its impersonal nature and lack of subjective bias, aligns well with the underlying rationale of the rule of law, as well as its objectives of accuracy, objectivity, and consistency. In an era where public tasks are becoming increasingly specialized and complex, AI offers a groundbreaking solution for efficiently addressing such challenges. However, AI also poses potential threats to the rule of law. The expansion of algorithm-driven regulation could diminish the relative status of individuals in relation to the state. Moreover, AI's technical characteristics may hinder transparency in governmental actions, the implementation of fair procedures, accountability, and the availability of effective remedies for individuals.
Even when AI is employed in the exercise of public authority, it must remain subject to the principles of the rule of law. In fact, AI’s integration should serve as an opportunity to advance the rule of law. To this end, it is essential to establish clear and concrete legal foundations for AI utilization in governance. Given AI’s opacity, the potential for misuse, and its significant implications for fundamental rights, a separate legal basis should be established specifically for AI-based public authority, in addition to the existing legal grounds for administrative actions. Moreover, the use of AI in public decision-making must ensure transparency, fairness, and accountability while adhering to due process requirements. The degree and manner in which these principles are implemented may vary depending on the nature of the governmental function and its impact on fundamental rights. Additionally, despite the increasing use of AI in public administration, mechanisms should be in place to guarantee human oversight and discretionary judgment. The extent of human involvement should be determined based on a comprehensive assessment of factors such as the nature and severity of AI-driven decisions affecting individual rights and obligations, the necessity and suitability of AI utilization, and the existence of procedural safeguards ensuring fairness and transparency.