Anthropogenic emissions alter atmospheric composition and therefore the climate, with implications for air pollution- and climate-related human health. Mortality attributable to air pollution and non-optimal temperature is a major concern, which is subject to change in the future under different climate change and socioeconomic scenarios. The goal of the study is to assess future changes in mortality attributable to long-term exposure to both non-optimal temperature and air pollution simultaneously, using model outputs from the recent Intergovernmental Panel on Climate Change multi-institution simulations. We show that, under a moderate scenario (SSP2-4.5), end-of-century mortality could quadruple from present-day values to around 30 (confidence level 95%:12-53) million/year. While pollution-related mortality is projected to increase five-fold by the end of the century in a moderate scenario, temperature-related mortality will experience a seven-fold rise. This increase causes non-optimal temperature to become a more important health risk factor than air pollution for at least 20% of the world's population. Population aging emerges as the primary driver of increased mortality, countering efforts to improve air quality and mitigate climate change. These findings underscore the urgency not only to improve air quality but, more importantly, to simultaneously implement more effective climate change policies. This will prevent significant loss of life in the future, which could offset the gain obtained by air quality improvements.