Flash drought (FD) is a type of extreme events characterized by rapid intensification of drought/dryness conditions, severely affecting vegetation growth and agricultural production. However, how the spatiotemporal dynamic risk of FD will change under future climate warming remains unclear owing to the limitations in the identification methods of FD dynamic characteristics and large uncertainty in global climate model projections. Here, we propose a comprehensive framework for projecting spatiotemporal dynamic characteristics of FD under future climate change by coupling the CMIP6 multi-GCM ensembles, the Variable Infiltration Capacity model, and an extended three-dimensional (3D) FD identification method. Based on this framework, we assess the projected changes in 3D dynamic characteristics of FD in China under SSP126, SSP245 and SSP585 scenarios. Model projections indicate a significant influence of climate change on FD characteristics, but the effects on different FD characteristics vary significantly. Climate change is expected to exacerbate FD severity and intensity for almost 75.3% of the national total area in China under three SSP scenarios, and prolong FD duration by 4.2%similar to 9.7%, especially under the higher SSP scenarios. Moreover, the migration distance of long-duration (>2-month) FD events tends to extend over time under all SSP scenarios, leading to an expansion of FD impact area in China under SSP245 and SSP585 (<36.4%). However, climate change will not alter the overall migration direction of FD, mostly from the humid South China to the arid and semi-arid North China regions. Our results emphasize the great importance of understanding the climate warming impacts on FD from a multi-attribute perspective.