Diamond is an important material for extreme-condition applications, yet its atomic-scale damage mechanisms under strong shock remain unclear. In this work, a carbon machine-learned potential for extreme conditions was developed within the neural evolution potential (NEP) framework, and large-scale molecular dynamics simulations were performed to investigate the dynamic response of single-crystal diamond under different crystallographic orientations and shock intensities. The results show pronounced crystallographic anisotropy under strong shock loading. Significant differences are observed among the [100], [110], and [111] orientations in elastic-plastic wave structures, stacking-fault evolution, and subsequent damage processes, with the [111] orientation exhibiting a more distinct two-wave structure. During shock unloading, diamond undergoes a continuous structural evolution from stacking faults or anomalous stacking to disordering, followed by graphitization and spallation. The [100] and [110] orientations tend to fracture through defect-assisted damage, whereas the [111] orientation is more prone to layered graphitization and early failure. These findings provide a reference for understanding and predicting the dynamic response of materials under extreme conditions.

Diamond,Shock,molecular dynamics