We validate the duration of ground‐motion time series produced by simulation routines implemented on the Southern California Earthquake Center broadband platform. This validation is based on comparisons of median and standard deviation of simulated durations and their trends with magnitude and distance, with semiempirical ground‐motion prediction equations (GMPEs). The GMPEs selected from the present work are based on a large database that includes five target events from California considered in the simulations. The duration parameters used for this validation are significant durations for 5%–75%, 5%–95%, and 20%–80% of the normalized Arias intensity. Some misfits are observed in the median and dispersion of durations from simulated motions and their trend with magnitude and distance. Understanding the source of these misfits can help guide future improvements in the simulation routines.