Water scarcity is becoming a critical global issue due to rapid population growth, climate change, and increasing urbanization. Efficient water resource management requires accurate forecasting of water demand to ensure sustainable supply and distribution. However, traditional forecasting methods often fail to capture complex, nonlinear, and dynamic water consumption patterns influenced by multiple factors. To address this limitation, this study utilizes Artificial Intelligence (AI) and Machine Learning (ML) techniques, specifically Artificial Neural Networks (ANN) and Long Short-Term Memory (LSTM) models, to improve prediction accuracy. The proposed models analyze historical water consumption data along with meteorological, socioeconomic, and temporal variables to forecast future demand more effectively. AI models are capable of identifying hidden patterns, seasonal variations, and long-term dependencies in water usage. This enables more precise and reliable prediction compared to conventional statistical approaches. The performance of the models is evaluated using standard accuracy metrics such as Mean Absolute Error (MAE) and Root Mean Square Error (RMSE), which demonstrate the effectiveness of AI-based methods. The results indicate that ANN and LSTM models significantly enhance forecasting accuracy and reliability. These predictive systems help water authorities optimize distribution, reduce wastage, and improve planning. Overall, AI-based water demand forecasting supports sustainable water management and ensures efficient utilization of water resources.