Water quality is vital for public health and ecosystem sustainability. However, approximately 2 billion people worldwide do not have access to drinking water that is appropriately managed, contributing to significant health risks and environmental issues. This paper proposes an efficient approach to classifying water quality as potable or impotable water using artificial neural networks (ANNs). This work highlights the perspective of ANNs in reducing water pollution and ensuring safe water access. The ANN is implemented with a dataset consisting of 2 water classes based on 3,276 water samples. The proposed ANN is tested via the Keras framework with a tuning technique to obtain the most significant water quality classification (WQC) precision. An error matrix is used to evaluate the ANN efficiency. The results show that the ANN is highly efficient in the classification of water quality. Moreover, The ANN provides a test accuracy of 68.38%, a mean precision of 69.64%, and the average area under the ROC curve for the implemented ANN is 51%.