CLOUD SECURITY ANOMALY DETECTION USING AI: A REAL TIME ANOMALY DETECTION SYSTEM FOR CLOUD WORKLOADS

Abstract

Cloud computing has been integrated into the core of contemporary computing infrastructures, and it has allowed organizations to effectively support large and dynamic workloads. Nevertheless, cloud environments have become more complex and diverse, posing serious security issues. Conventional intrusion detection systems, which make use of preset rules or known attack signatures, do not reveal the unknown or developing threats. In this paper, the researcher will present Intelli Guard Cloud, an artificial intelligence-based real-time anomaly detection system to improve the security of the cloud environment. The system continuously observes the workloads in the cloud and multi-source telemetry, such as log records of the system, network traffic, API activity, and metrics of resource use. It uses a hybrid detection model based on the combination of an Autoencoder model, an Isolation Forest model and a Long Short-Term Memory (LSTM) model to identify statistical, structural, and temporal anomalies. The suggested framework is tested on simulated and real-life cloud workload data under simulated attack conditions. The experimental findings point to the fact that the system delivers better detection accuracy, lower-false positive and low-latency response in comparison with the traditional approaches. These results prove how AI-based solutions can be effective in providing adaptive and scalable anomaly detection to dynamic clouds.