Malignant gliomas are the most common and aggressive primary brain tumor. A major barrier to the treatment of glioma is the drug resistance. The existence of glioma stem cells may be responsible for drug resistance. Recent clinical data show that arsenic trioxide ($As_2O_3$) causes remission in patients with acute promyelocytic leukemia without severe side effects. However, it is still unknown that whether $As_2O_3$ could inhibit the growth of glioma stem cells. Glioma stem cells are also called glioma stem/progenitor cells (GSPCs). The GSPCs are a population of cells capable of extensive self-renewal, differentiation into multiple lineages, initiated the original tumor in immunodeficient mice and often co-expressed differentiation surface marker with the stem/progenitor cell marker nestin. In this study, we cultivated GSPCs, which were established from the excised tumor tissue of a patient, and showed that $As_2O_3$ could inhibit the proliferation of GSPCs. Treatment of GSPCs for 24 h, 48 h and 72 h with $As_2O_3$ induced cell death in a dose- and time-dependent manner. Subsequently, we demonstrated that the $As_2O_3$-induced GSPCs death was due to cell cycle arrest at the G0/G1 transition and was associated with activation of c-jun N-terminal kinase (JNK) pathway. The findings underscore the potential of $As_2O_3$ to inhibit the proliferation of GSPCs through the JNK pathway and suggest its application to the effective therapy for patients with glioma.