The morphology of highly braided large alluvial stream like the Brahmaputra river is fraught with great deal of complexities. Due to the inherent intricacies involved, conventional approach for morphological modelling for a river like the Brahmaputra will not be suitable due to intricate relationships between the channel geometrical variables. In this study, an attempt has been made to apply the Artificial Neural Network (ANN) technique, to develop ANN model of Brahmaputra river to predict cross-section profile and top width. The interpolated and extrapolated cross-section profile in the past, present and future with respect to time as predicted by ANN models are found to be encouraging which merits more further study.