Accurate prediction of fatigue life has long been a critical challenge in the design and development of advanced equipment. In recent years, artificial intelligence (AI) models have introduced a new paradigm for fatigue life prediction, offering promising solutions. In particular, the integration of existing knowledge into AI models can significantly enhance their training and predictive capabilities. This paper provides a comprehensive review of the current state of research on knowledge-guided and data-driven models of fatigue life prediction, highlighting their respective strengths and weaknesses, and outlining fthe challenges and possible future perspectives.