In the face of increasing cyber threats, traditional signature-based intrusion detection systems (IDS) are struggling to keep up with the evolving sophistication of cyber-attacks. The need for more adaptive and accurate detection techniques has led to the integration of Artificial Intelligence (AI), particularly Machine Learning (ML), Deep Learning (DL), and ensemble learning approaches. This paper explores the use of Graph Neural Networks (GNN) for enhancing Network Intrusion Detection Systems (NIDS), particularly focusing on the CICIDS2017 dataset. By employing AI-based models, this research highlights the potential of GNNs in accurately classifying multi-class cyber-attacks. The study provides an overview of various AI techniques, including GNN and Random Forest, and their applications in NIDS. The goal is to address the challenges faced in multi-class attack detection and offer future insights for advancing intrusion detection systems in complex and dynamic network environments.

Keywords: Artificial Intelligence, Machine Learning, Deep Learning, Graph Neural Networks, Random Forest, Intrusion Detection, Cyber-attacks, Multi-classification, Network Security, CICIDS2017 dataset.