AbstractThe net effective temperature (NET), an index that includes the combined effects of temperature, humidity, and wind, was used along with temperature to assess the impacts of climate change on the heat stress perception in East Asia, one of the regions considered most vulnerable to heat stress. The need for dynamic downscaling has been emphasized because the regional effects of climate change do not follow the global levels linearly. In this study, daily maximums calculated from the 3‐hourly data downscaled by five different regional climate models from four coupled general circulation models participating in Coordinated Regional Climate Downscaling Experiment‐East Asia phase 2 were utilized. To account for the fact human beings acclimate to their environments, 95th percentile of the maximum temperature and maximum NET was used along with the average boreal summer maximum temperatures/NETs. The performance of the models was assessed first, which showed that the models reproduced the current climate well. Future projections revealed an increase in both average and 95th percentile of the maximum temperature and NET over the entire domain for both the RCP8.5 and SSP5‐8.5 scenarios. The increase in heat stress (NET) was slightly larger than the temperature itself, with an increase of up to 7/10°C for temperature and 8/11°C for NET in RCP8.5/SSP5‐8.5, respectively. The overall increases in temperature and NET were projected to be higher in the higher latitudes, while the increase in the frequency of the temperature and NET extremes was predicted to be higher in the already vulnerable regions in the southern part of the domain.