We study the impact of short-term exposure to ambient air pollution on the spread and severity of COVID-19 in Germany. We combine data on county-by-day level on confirmed cases and deaths with information on local air quality and weather conditions and exploit short-term variation in the concentration of particulate matter (PM10) and ozone (O3). We apply fixed effects regressions controlling for global time-varying confounding factors and regional time-invariant confounding factors on the county level, as well as potentially confounding weather conditions and the regional stage of the pandemic. We find significant positive effects of PM10 concentration after the onset of the illness on COVID-19 deaths specifically for elderly patients (80+ years): higher levels of air pollution by one standard deviation 3 to 12 days after developing symptoms increase deaths by 30 percent (males) and 35 percent (females) of the mean. In addition, air pollution raises the number of confirmed cases of COVID-19. The timing of results supports mechanisms of air pollution affecting the severity of already realized infections. Air pollution appears not to affect the probability of infection itself.