In today's the current power distribution networks, electricity theft poses challenges to power reliability and creates significantly large amounts of economic losses. With the implementation of smart meters and the widespread installation of Internet of Things devices, it is necessary to utilize intelligent monitoring strategies for the purpose of detecting abnormal energy usage in real-time. This paper discusses an Intelligent Power Monitoring and Electricity Theft Detection System utilizing the Random Forest machine learning algorithm. RMS feature extraction produces a large amount of data, which is then classified into normal or abnormal electricity consumption patterns using a Random Forest classifier. The Random Forest algorithm is an ensemble classifier that provides accurate classification of the data without overfitting and with low false-positive detection rates. The system communicates in real-time via an IoT communication framework and initiates control once theft detection occurs.