Abstract: Wave energy derived from oceanic kinetic forces is a highly promising renewable energy source. As global efforts to incorporate renewable energy into the grid increase, accurate wave energy forecasting becomes essential for optimizing energy harvesting and grid integration. This paper examines the latest developments in machine learning (ML) approaches, focusing on deep learning (DL), ensemble methods, and hybrid models used for forecasting ocean wave energy. It highlights the strengths and weaknesses of various approaches in capturing the complex nonlinear dynamics of ocean waves, including predicting energy flux, significant wave height (SWH), and wave period. Additionally, the paper explores how hybrid models, combining physical models with ML, have emerged as powerful tools for improving forecast accuracy over traditional methods. This review concludes with insights into future directions, emphasizing the potential of advanced techniques like transformers, generative adversarial networks (GANs), and real-time data assimilation for enhancing prediction reliability and computational efficiency.