형사사법정보의 개인정보 보호와 활용 균형 방안 연구 - 비식별화 기술로서 재현 데이터를 중심으로 -

This study explores the balance between efficient utilization of criminal justice information and personal data protection in the Fourth Industrial Revolution. Criminal justice information is vital for crime prevention and investigation but its sensitivity creates a dilemma, conflicting with privacy laws, criminal procedure principles (due process, presumption of innocence), and informational self-determination rights. Limitations of existing de-identification technologies and difficulties in obtaining data subject consent are major constraints. Thus, this study proposes the integrated application of synthetic data and de-identification technologies as a core solution. Synthetic data, maintaining original data's statistical properties without personal identifiers, offers an innovative alternative enabling AI model training and advanced analysis without sensitive information leakage. This research analyzes KICS data management and privacy issues, detailing synthetic data's concept and specific advantages in criminal justice. Furthermore, it examines domestic and international cases of synthetic data and de-identification technology, confirming their contribution to data accessibility, privacy protection, policy support, and AI development. Based on this, practical strategies are suggested to minimize side effects during synthetic data adoption, including pilot implementation, phased expansion, risk verification, and continuous feedback system establishment. In conclusion, this study demonstrates that the integrated application of synthetic data and de-identification technologies enables safe and efficient use of criminal justice information, maximizing crime prevention and investigation efficiency while actively protecting citizens' right to informational self-determination. This will be key to advancing smart policing systems in the Fourth Industrial Revolution.