This study investigates the warm-season extreme precipitation–temperature scaling relationship and its implications on climate change based on CONUS404, a dynamical downscaling product from convection-permitting Weather Research and Forecasting (WRF) model simulations over the conterminous US for the past four decades. In comparison with CONUS404, we also analyze the historical simulation from the WRF-Thermodynamic Global Warming (WRF-TGW) dataset, which is another dynamical downscaling product driven by atmospheric reanalysis. To capture the temporal variation of precipitation intensity, we use hourly data to examine how extreme precipitation intensity (EPI) varies with temperature and atmospheric saturation deficit over representative regions in the United States and focus on the saturated condition under which EPI tends to monotonically increase with temperature. Relative to NASA’s Integrated Multi-satellitE Retrievals for the Global Precipitation Measurement (IMERG) data (version 6), WRF generally overestimates the positive scaling rates in saturated atmosphere, a model signature that is robust against model spatial-resolution (12km vs. 4km) and whether the model resolves or parameterizes deep convection. Moreover, a comparison of EPI–temperature scaling rates between the weather and decadal timescales produces a positive correlation, indicating that the weather-scale metrics may serve as a potential emergent constraint for climate-induced changes. Additional analyses are being conducted based on NCEP Stage-IV data to assess the uncertainties associated with precipitation observations.