AI-Driven Vulnerability Management Framework for the Internet of Things
TL;DR
This study proposes an AI driven vulnerability management framework that integrates anomaly detection techniques using machine learning to enhance proactive threat identification and mitigation in IoT ecosystems, and demonstrates that AI based models significantly outperform traditional methods in detecting previously unseen threats.
Abstract
This rapid growth of IoT has changed the landscape of today’s digital world by allowing devices to communicate effectively, especially in different fields like healthcare, smart cities, industrial control, and defense. Despite its advantages, IoT introduces significant security challenges due to device heterogeneity, constrained computational resources, and weak security architectures, making it highly vulnerable to cyber threats. Traditional vulnerability management approaches, including rule-based intrusion detection systems and signature-based scanning, have proven inadequate in addressing the dynamic and large-scale nature of IoT environments, as they are largely reactive and incapable of detecting novel attack patterns. This study proposes an AI driven vulnerability management framework that integrates anomaly detection techniques using machine learning to enhance proactive threat identification and mitigation in IoT ecosystems. The framework leverages publicly available datasets such as Bot-IoT, CIC-IoT, and UNSW NB15 to train and evaluate models capable of distinguishing between normal and malicious network behaviors. Various machine learning algorithms, including supervised and unsupervised techniques, were implemented and assessed using performance metrics such as accuracy, precision, recall, F1-score and false positive rate. The results demonstrate that AI based models significantly outperform traditional methods in detecting previously unseen threats, achieving high detection accuracy and reduced false positives. The proposed framework integrates anomaly detection into a structured vulnerability management lifecycle encompassing identification, prioritization and remediation of vulnerabilities. Generally, the study provides a scalable and adaptive solution for improving IoT security, reducing system vulnerabilities, and enhancing resilience against evolving cyber threats, with potential for future real-world deployment across critical sectors.