This study explores an application of machine learning (ML) to predict the insurance premiums for life insurance applicants. Traditional methods often face challenges in accurately assessing risk due to the complexity and diversity of applicant data. By leveraging ML algorithms, we analyze a wide range of applicant attributes, including demographic details, medical history, lifestyle factors, and financial indicators. In this research, we employ models such as linear regression (LR), decision trees (DT), random forests (RF), and gradient boosting (GB), with feature engineering techniques used to enhance the accuracy of predictions. Our findings indicate that ML models significantly outperform traditional methods in terms of risk assessment and pricing strategies for insurers. Specifically, the LR model achieved an accuracy of 74%, demonstrating moderate performance. The RF regressor, with an accuracy of 83%, exhibited better predictive capabilities, particularly in handling complex datasets and capturing nonlinear relationships. GB, however, outperformed all other models with an accuracy of 86%, showcasing its strong predictive power. The ability of gradient boosting to iteratively improve the performance of weaker models contributed to its superior results. This research advances underwriting practices by demonstrating the potential of ML to revolutionize risk assessment and pricing strategies in the insurance industry, to provide precise and data-driven decision-making tools for insurers.