Pipelines are essential infrastructure for transporting oil and gas, but leaks might possess devastating consequences, including fires, injuries, property damage and environmental pollution. Thus, ensuring pipeline integrity is essential for a safe and sustainable energy supply. This paper explores the application of Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL) to enhance the prediction of leaks in oil and gas pipeline systems, which is vital for ensuring environmental safety and economic stability. Through a comprehensive review and data-driven methodologies, the study demonstrates how ML algorithms, including neural networks and deep learning models, significantly outperform traditional leak detection methods in terms of accuracy. The research introduces anomaly detection models driven by machine learning and deep learning to address the issue of leakage in oil and gas pipelines. Our contribution is particularly distinguished by applying deep learning techniques, specifically Multi-Layer Perceptron (MLP) and Convolutional Neural Network (CNN), to numerical data from the oil and gas sector. In addition, various ML algorithms were employed, such as K-Nearest Neighbor, Decision Tree, Random Forest, Support Vector Machine (SVM), and Gradient Boosting, to create reliable models for predicting pipeline leaks. Among these, the SVM algorithm achieved an accuracy of \(96\%\) , notably outperforming other models. This study underscores the potential of these technologies to provide timely and precise leak detection, ultimately contributing to better management and maintenance of critical infrastructure.