Traditional risk assessments focus on individual hazards in a specified geographic area during a given period of time, thereby disregarding the complex interplay between different hazards. It will, therefore, underestimate the risks associated with extreme conditions. Drought and heatwave are two of the deadliest and costliest natural hazards which often occur concurrently or subsequently, creating compound and cascading extreme events. Their risks under future socio-economic developments and climate change must, therefore, be analyzed in a compound manner to provide more reliable information for disaster risk reduction efforts and adaptation policies. This research will be the first attempt to quantify, at the global scale, the change in risk of compound/cascading drought and heatwave events. Big climate model data will initially be processed using machine learning techniques. Drought and heatwave hazards will then be characterized by a variety of indicators and the drought-heatwave interaction will be modelled using the joint distribution probability. The risk will finally be projected towards the end of the twenty-first century as a function of hazard, exposure, and vulnerability under four emission and socio-economic scenarios.