경찰의 AI CCTV 활용에 대한 시민의 수용 의사

초록

As the paradigm of contemporary policing shifts toward Scientific Policing, AI CCTV serves both as a tool for enhancing crime prevention efficiency and as a mechanism that repositions citizens as subjects of surveillance, creating a unique tension. This study focuses on citizens’ acceptance intention as a critical variable determining the successful adoption of such technology, and tests an extended Technology Acceptance Model (TAM) that incorporates ‘trust in AI-based policing’ as a key antecedent. Notably, given that citizens are not active operators but rather passive targets of the technology, the conventional concept of perceived ease of use was reinterpreted as cognitive accessibility-the degree to which citizens can readily understand and comprehend the technology’s operational principles. Survey data collected from 1,101 adults nationwide were analyzed using Structural Equation Modeling (SEM) to examine influence pathways and mediating effects. The results indicated that trust in AI-based policing had a significant effect on both perceived usefulness and perceived ease of use, and all mediating pathways from trust to acceptance intention were statistically significant. Perceived usefulness emerged as the most powerful predictor of acceptance attitude and intention. This suggests that citizens place greater weight on the technology’s tangible contribution to community safety than on its procedural convenience. This study is significant in that it empirically demonstrates that institutional trust serves as a decisive variable in mitigating perceived risks and facilitating technology acceptance for public safety technologies directly linked to fundamental rights. Future deployment of AI-based policing technologies should prioritize building public trust through operational transparency and institutional safeguards alongside technological advancement.

키워드