ARCHIVES

Cloud computing is an essential part of the contemporary digital environment due to its scalability and flexibility. However, the migration to cloud platforms introduces some substantial cybersecurity challenges, among which unauthorized access, data breaches, distributed denial-of-service (DDoS) attacks, and insider threats. The traditional cybersecurity measures, including rule-based firewalls and signature-based intrusion detection systems, are not efficient enough in the modern digital environment. This paper proposes an AI-based threat detection framework that utilizes cloud log monitoring, classification, and anomalous behavior detection to identify malicious activities in cloud infrastructure. The conceptual framework was evaluated on CICIDS2017 and UNSW-NB15 public intrusion detection datasets using various machine learning (ML) algorithms. The performance of Logistic Regression, Support Vector Machine, Random Forest, and Gradient Boosting classifiers was tested, and the accuracy of the best-performing model, Random Forest, reached 96.8%. The proposed framework includes a risk prioritization module and automated response mechanisms to ensure quick threat mitigation and enhance cloud security. The study concluded that an AI-based threat detection framework could be beneficial for the cybersecurity infrastructure by providing improved protection against network attacks and enabling dynamic adaptive defenses.