Generative AI for Real-Time Cloud Security: Advanced AnomalyDetection Using GPT Models

Abstract

Cloud infrastructures are becoming
increasingly complex, creating a strong need for
advanced real-time security solutions. Traditional
anomaly detection systems, which rely on
predefined rules and signature-based methods,
often fail to detect novel and sophisticated cyber
threats. This paper explores the use of generative AI
models, including LLaMA and GPT architectures,
to enhance cloud security through real-time
anomaly detection. The proposed framework
leverages the pattern recognition and adaptive
learning capabilities of these models to analyze
large-scale cloud data such as logs, network traffic,
user behavior, and system activities. By
continuously learning from new data, the system
improves its ability to identify previously unknown
threats. This approach addresses key limitations in
existing cloud security methods by introducing a
scalable and adaptive solution. The use of
generative AI enables more accurate and timely
detection of anomalies in dynamic cloud
environments. Furthermore, the framework
demonstrates how GPT-based models can generate
meaningful insights from diverse inputs. The study
highlights the potential of integrating generative AI
into cloud security systems for enhanced protection.
It also evaluates the feasibility of deploying such
models in large-scale infrastructures. Overall, the
research emphasizes improved resilience,
adaptability, and efficiency in modern cloud
security frameworks.