The Use of Artificial Intelligence in Crime Prediction and Prevention for Public Safety

Today, artificial intelligence(AI) is rapidly emerging as a core technology for ensuring public safety, fundamentally reshaping the traditional concept of policing centered on the prediction and prevention of crime. In particular, AI-based predictive policing systems and intelligent CCTV are being actively implemented in countries such as South Korea and the United States, shifting the paradigm of law enforcement from reactive investigations to real-time surveillance and proactive intervention. In Korea, AI policing technologies are being increasingly applied in areas such as protection of stalking victims, recidivism prediction for individuals under electronic monitoring, and crowd density analysis. However, the legal and ethical frameworks to regulate these technologies remain insufficient. This paper examines the global trends of AI-based policing technologies alongside Korea’s policy and legislative developments, while focusing on the potential for human rights violations and discriminatory outcomes as AI autonomy expands under the banner of public safety. Issues such as algorithmic bias, lack of transparency, and the ambiguity of accountability pose serious challenges, particularly in the field of law enforcement. Without securing fairness in targeting and clarity in the attribution of responsibility, the use of predictive models may violate constitutional rights and procedural due process. Ultimately, it is essential to develop legal and policy safeguards in step with the rapid advancement of technology. When AI policing technologies are applied, efforts must be made to strike a balance between the public interest of crime prevention and private interests such as privacy and equality, avoiding the sacrifice of one value for another. In designing policing strategies in the era of artificial intelligence, it is necessary to fully utilize the benefits of technology while ensuring that such systems operate within the boundaries of the rule of law and the protection of fundamental rights. Accordingly, this paper emphasizes that future-oriented policing models employing AI for public safety must be built on the principle of checks and balances, allowing AI to enhance social security as a tool for humans without infringing upon individual freedoms and rights.