Abstract: Electrochemical energy storage technologies, represented by lithium-ion batteries, are significant supporting technologies and key equipment for building new energy vehicle and novel power systems. However, safety accidents occur frequently in lithium-ion battery power and energy storage systems, mainly due to safety failures during its operation, resulting in thermal runaway and further catastrophic consequences such as fires or explosions. Therefore, it is necessary to achieve timely and accurate active safety warning before the failure of lithium-ion batteries, to avoid battery thermal runaway and ensure the safe operation of power and energy storage battery systems. The research focuses of four types of early active safety warning methods for thermal runaway of lithium-ion batteries based on signal characteristics, model prediction, data-driven, and hybrid strategies are systematically summarized. The theoretical basis and implementation paths of each method are introduced, and the advantages and disadvantages between different methods are analyzed and described. Finally, the key issues yet to be resolved in the field of active safety warning for lithium-ion batteries are summarized, and possible solutions are proposed, providing valuable references for subsequent related research.