Competition over limited water resources is one of the main concerns for the next future. In this paper, we presented an innovative analysis of the past hydro-political interactions (conflict and cooperation over water resources) in international river basins and their determinants through the application of the Random Forest algorithm. Our aim was twofold: on the one hand, we aimed at highlighting the factors that are more relevant in determining the water cross-boundary issues, capturing also the non-linear relations between the main drivers. On the other hand, we aimed at producing a spatially explicit data driven indicator able to map and monitor the evolution of the likelihood of hydro-political interactions over space and time, under changing socio-economic and bio-physical scenarios. Historical (1997-2007) international water cross-border issues, both classified as conflict or cooperation, were put in relation with information about river basin freshwater availability; climate stress; human pressure on water; socio-economic conditions, including institutional development and power imbalance; and topographic characteristics. The application of the Random Forest approach outperformed alternative methodologies, such as linear models and statistical learning algorithms (RMSE = 0.218, R2=0.68). The analysis allowed to identify the most important factors determining water issues, such as water availability, population density, power imbalance, and climatic stressors, estimating their non-linear marginal impact on the likelihood. The model fitted on historical observations was used to estimate a baseline condition of the spatial distribution of the likelihood of hydro-political interaction around the globe at high spatial resolution. This baseline scenario was then compared to four future climate and population density scenarios aimed at estimating the hydro-political risk under future (2050 and 2100) conditions considering two greenhouse gases emission scenarios (RCP 4.5 and 8.5) from five different downscaled and bias corrected General Circulation Models (GCMs) in a multi-model mean. The result of this work allows to identify current and future areas where water issues are more likely to rise, and cooperation over water should be pursued to avoid possible tensions especially under changing environmental conditions.