This study reviews the role of artificial intelligence (AI) in coronary artery calcium scoring (CACS) for coronary heart disease (CHD) risk assessment. Current CACS practices rely on dedicated CT protocols, which are time-consuming and contribute to radiation exposure. The study explores AI's potential to improve CACS efficiency and repurpose non-dedicated CT scans, which could reduce costs and radiation exposure but face challenges like motion artifacts. The review discusses the development of automated CACS using deep learning (DL) algorithms, specifically convolutional neural networks (CNNs), and evaluates their performance across various CT protocols. It highlights the benefits of AI in automating calcium segmentation and quantification, thereby reducing manual workload and increasing accuracy. However, barriers such as data diversity, external validation, and noise reduction need to be addressed. The study emphasizes the need for large, diverse datasets and collaboration to enhance AI model generalizability and clinical implementation.