Charged inverse micelles (CIMs) generated during a continuous polarizing voltage between electrodes in the model system of polyisobutylene succinimide in dodecane do not populate a diffuse double layer like CIMs present in equilibrium (regular CIMs), but instead end up in interface layers. When the applied voltage is reversed abruptly after a continuous polarizing voltage step, two peaks are observed in the transient current. The first peak is due to the release of regular CIMs from the diffuse double layers formed during the polarizing voltage step, which is understood on the basis of the Poisson–Nernst–Planck equations. The second peak is due to the release of a small fraction of generated negative CIMs from the interface layer. A model based on space charge limited release of the generated negative CIMs from the interface layer is presented and the results of the model are compared with several types of measurements. For the situation in which the bulk is deprived of regular CIMs and neutral inverse micelles, the results of the model are in agreement with the experimental results. However, for the situation in which regular CIMs and neutral inverse micelles are present, the model shows discrepancies with the experiment for high voltages and high charge contents. These discrepancies are attributed to electrohydrodynamic flow caused by local variations in the electric field at the vicinity of the electrodes, which occur during the reversal voltage. Also the long term decrease of the amount of released generated CIMs is studied and it is found that the presence of regular CIMs and neutral inverse micelles speeds up the decrease. This study provides a deeper insight in the electrodynamics of CIMs and is relevant for various applications in non-polar liquids.