In today’s rapidly evolving technological world, cyber threats are growing more complex and are getting hard to detect. The organizations, to detect, avoid or thwart cyber threats, rely on various methods, tools, frameworks, and among those intrusion detection systems (IDS) being an important, widely used tool to identify anomalies in the network. The traditional IDS and machine learning-based IDS also fall short lacking accuracy in detecting or preventing attacks. So the study explores the application of Generative Adversarial Network to enhance IDS performance. By generating adversarial network traffic that mimics legitimate behavior, the proposed framework helps test and improve the robustness of IDS, paving way for more resilient security solutions.