As computers and smart devices become more integrated into daily life and business, computer networks face increased security risks. Traditional firewalls are insufficient against sophisticated attacks, making Network Intrusion Detection Systems (IDS) crucial for detecting and raising alarms about network threats. IDS typically rely on predefined attack databases, which can fail against novel (0- day) attacks. To address this, researchers have explored machine learning and deep learning for anomaly detection. This study proposes an AI model using 2D Convolutional Neural Networks (CNN) to detect network anomalies and alert to new intrusions. The approach includes converting tabular data into 2D images using the Image Generator for Tabular Data (IGTD) algorithm, maintaining and enhancing the understanding of feature relationships. The model’s performance is tested on the NSL-KDD dataset and incorporates AI explainability for better analysis.