The paper addresses the significant issue of electricity theft in smart grids, which causes substantial financial losses for power utilities. Traditional methods for detecting electricity theft face challenges such as the curse of dimensionality and highly imbalanced data distribution. To overcome these issues, the authors propose a hybrid system that combines Multi-Layer Perceptron (MLP) and Gated Recurrent Units (GRU). The system is applied to data from the Chinese National Grid Corporation (CNGC) and involves data preprocessing, balancing using k-means SMOTE, and applying both GRU and MLP models. The performance of the proposed system is evaluated using various metrics such as accuracy, F1-score, precision, and recall. The results show that the proposed hybrid system outperforms other models like Alexnet, GRU, BGRU, and RNN in terms of accuracy and efficiency. The study also includes a comparison with existing literature, highlighting the superiority of the proposed method. The conclusion emphasizes the effectiveness of the proposed hybrid system in detecting electricity theft in smart grids.The paper addresses the significant issue of electricity theft in smart grids, which causes substantial financial losses for power utilities. Traditional methods for detecting electricity theft face challenges such as the curse of dimensionality and highly imbalanced data distribution. To overcome these issues, the authors propose a hybrid system that combines Multi-Layer Perceptron (MLP) and Gated Recurrent Units (GRU). The system is applied to data from the Chinese National Grid Corporation (CNGC) and involves data preprocessing, balancing using k-means SMOTE, and applying both GRU and MLP models. The performance of the proposed system is evaluated using various metrics such as accuracy, F1-score, precision, and recall. The results show that the proposed hybrid system outperforms other models like Alexnet, GRU, BGRU, and RNN in terms of accuracy and efficiency. The study also includes a comparison with existing literature, highlighting the superiority of the proposed method. The conclusion emphasizes the effectiveness of the proposed hybrid system in detecting electricity theft in smart grids.