GenAI Based Identification and Analysis of Security Risks and Vulnerabilities in 5G Network Traffic

5G networks become integral to modern communication which ensure their security against emerging threats has become a critical challenge. This research investigates the security risks and vulnerabilities in 5G network traffic to focus on the comparative performance of traditional machine learning (ML) models and generative artificial intelligence (GAI) techniques for attack detection. Specifically, the study evaluates the detection accuracy of DoS, MITM, and DDoS attacks across both traditional ML and GAI models. The findings reveal that GAI significantly outperforms traditional ML models in terms of detection accuracy with an average improvement of 15-20%. The study also explores the potential privacy and performance trade-offs associated with each approach. The results show that while generative AI introduces a slight increase in latency compared to traditional models, the improved security benefits justify this trade-off. This research highlights the promising role of GAI to enhance the security and privacy of 5G networks which offer a robust solution to the evolving threats in next-generation communications. The study concludes by recommending for further exploration into hybrid models and real-time attack prediction to strengthen the security framework of 5G networks.