Abstract To characterize future drought events over a drought prone area like South Asia is paramount for drought risk mitigation. In this paper, a five-model ensemble mean from CMIP6 was chosen to project drought characteristics in South Asia under the latest SSPs-RCPs emission scenarios (SSP1–2.6, SSP2–4.5, and SSP5–8.5) for the period 2020–2099. Additionally, corresponding scenarios RCP2.6, RCP4.5 and RCP8.5 of CMIP5 were used for comparison and to identify the changes and improvements of CMIP6 over the South Asia. Principle Component Analysis and the Varimax rotation method is used to divided the study area into five homogenous drought sub-regions. Drought duration, frequency, and intensity are analyzed based on the Run theory, and the Standardized Precipitation Evapotranspiration Index (SPEI) at 12-months timescale, and the self-calibrating Palmer Drought Severity Index (sc-PDSI). The Modified Mann-Kendall and Sen's slope method is adopted to detect sub-regional trends in drought characteristics. Results show that significant increases in drought conditions mainly pronounced over the North-West sub-region. Strong increases are projected in the average drought duration and drought frequency. The North-West sub-region is the most vulnerable to face frequent drought events with longer duration with higher intensity. Parts of the South-West, North-Central, and North-East sub-regions will also face more adverse drought conditions in future. The selected model ensemble from CMIP6 has a very robust capability to simulate present climate parameters (precipitation, temperature, and evaporation) and satisfactorily captures drought characteristics in South Asia. These results provide a basis for developing drought adaptation measures for South Asia.