Ongoing human development into fire-prone areas contributes to increasing wildfire risk to human life. It is critically important, therefore, to have the ability to characterize wildfire risk to populated places, and to identify geographic areas with relatively high risk. A fundamental component of wildfire risk analysis is establishing the likelihood of wildfire occurrence and interaction with social and ecological values. A variety of fire modeling systems exist that can provide spatially resolved estimates of wildfire likelihood, which when coupled with maps of values-at-risk enable probabilistic exposure analysis. With this study we demonstrate the feasibility and utility of pairing burn probabilities with geospatially identified populated places in order to inform the development of next-generation, risk-based Wildland-Urban Interface (WUI) maps. Specifically, we integrate a newly developed Residentially Developed Populated Areas dataset with a stochastic, spatially-explicit wildfire spread simulation model. We classify residential population densities and burn probabilities into three categories (low, medium, high) to create a risk matrix and summarize wildfire risk to populated places at the county-level throughout the continental United States. Our methods provide a new framework for producing consistent national maps which spatially identifies the magnitude and the driving factors behind the wildland fire risk to populated places. This framework advances probabilistic exposure analysis for decision support in emergency management, rural and urban community planning efforts, and more broadly wildfire management and policy-making.