International Research Journal of Education and Technology

This paper examines the integration of artificial intelligence (AI) techniques to enhance Internet of Things (IoT) security frameworks in smart city environments. As urban centers increasingly deploy interconnected IoT systems to improve infrastructure management and public services, they simultaneously create expanded attack surfaces for cyber threats. We propose a layered security architecture that leverages machine learning algorithms to detect anomalies, predict potential vulnerabilities, and autonomously respond to security incidents across distributed IoT networks. Our experimental implementation demonstrates significant improvements in threat detection speed (47% faster than traditional methods) and reduction in false positives (68% decrease) when applied to smart traffic management systems. The findings suggest that AI-driven security approaches can provide the scalability and adaptability required to protect the complex, heterogeneous IoT ecosystems that underpin modern smart cities while minimizing human intervention requirements.