규제된 자율규제 형태로서 인공지능 표준(AI Standards)의 규범적 쟁점과 과제

With the advancement of Artificial Intelligence(AI) technology, concerns over its risks and social impact have intensified, highlighting the need for an effective regulatory framework. In response, countries around the world are establishing governance structures to ensure the safety and trustworthiness of AI systems, and standards are emerging as a core regulatory instrument within this context. AI standards, in particular, serve not only technical and industrial purposes but also function as regulatory tools. Within the framework of soft law—such as self-regulation and co-regulation—standards can achieve normative effectiveness, raising important questions for legal analysis from a regulatory law perspective. This paper examines the legal nature and normative function of standards as regulatory mechanisms for AI governance. It begins by identifying the limitations of traditional regulatory models and explores the theoretical structure and normative justification of standards, focusing on self-regulation and co-regulation frameworks. The study further analyzes how global standards function as regulatory tools through case studies from the United States, the European Union, and Germany. In addition, the paper reviews how standards are applied in South Korea, particularly under the Basic Act on Artificial Intelligence, and proposes a legal and institutional blueprint for the enactment and operation of AI standards. Ultimately, this paper argues that AI standards go beyond their technical function and can operate as preemptive regulatory norms, tools for inducing self-regulation, and mechanisms for implementing co-regulation. In doing so, it seeks to provide a legal roadmap for enhancing the regulatory coherence and effectiveness of AI governance.