Background Alternative splicing (AS), one of the main post-transcriptional biological regulation mechanisms, plays a key role in the progression of glioblastoma (GBM). Systematic AS profiling in GBM is limited and urgently needed. Methods TCGA SpliceSeq data and the corresponding clinical data were downloaded from the TCGA data portal. Survival-related AS events were identified through Kaplan-Meier survival analysis and univariate Cox analysis. Then, splicing correlation network was constructed based on these AS events and associated splicing factors. LASSO regression followed by multivariate Cox analysis was performed to validate independent AS biomarkers and to construct a risk prediction model. Enrichment analysis was subsequently conducted to explore potential signaling pathways of these AS events. Results A total of 132 TCGA GBM samples and 45,610 AS events were included in our study, among which 416 survival-related AS events were identified. An AS correlation network, including 54 AS events and 94 splicing factors, was constructed, and further functional enrichment was performed. Moreover, the novel risk prediction model we constructed displayed moderate performance (the area under the curves were > 0.7) at both one, two and three years. Conclusions Survival-related AS events may be vital factors of both biological function and prognosis. Our findings in this study can deepen the understanding of the complicated mechanisms of AS in GBM and provide novel insights for further study. Moreover, our risk prediction model is ready for preliminary clinical applications. Further verification is required.