Dynamic voltage stability deals with the voltage levels and how they’re affected by either faults or load changes within the system. Voltage instability has long been suspected in the voltage collapse and islanding of power systems. Identifications of operational conditions leading to voltage collapse is therefore critical in allowing for critical defensive measures by the system operator to avoid voltage collapse before it occurs. This paper examines the use of Voltage Collapse Proximity Indicator (VCPI) in conjunction with Artificial Neural Networks to predict conditions of voltage instability before they occur for load buses within the Kenya power system that can be used for online prediction of voltage stability within the system. The results show the validation of the VCPI index as an indicator of voltage stability when compared to previous studies on both the IEEE 9-bus system and IEEE 14-bus system and as an indicator of the point of voltage collapse. The application of Artificial Neural Networks to map the relationship between the power demand at a bus, total power demand in the system and the VCPI index to the Kenyan system shows a high level of accuracy leading to the conclusion that an online VCPI index prediction using an ANN can be used to predict voltage stability in the Kenya power system.