In recent years, Artificial Neural Networks (ANNs) have stood out among machine learning algorithms, being successful in a huge range of applications, especially in recognizing image, audio and video patterns, as well as in natural language analysis. The use of activation functions plays a crucial role in the operation of these algorithms, directly influencing the representation capacity and training effectiveness of ANNs. The logistic (or sigmoid) function is often used as a standard activation function in many neural network models due to its favorable properties of non-linearity and smooth derivatives. However, the existing literature lacks in-depth investigations into the potential of the Skew-Logistic (SL) function as a viable alternative, especially in scenarios where asymmetry in the data is a common reality. This work aims to investigate the SL function as an activation function in ANNs, exploring its ability to deal with asymmetric data. To achieve this, the SL function was implemented computationally in different neural network architectures. The models were trained on various databases selected for the experiments, and their performance was evaluated using standard metrics such as accuracy, precision, recall and F1-score. This procedure was carried out in each experiment with the SL and sigmoid activation functions in order to compare them. The results indicate that SL can bring improvements to the models in some asymmetric data sets, in which a significant increase in performance metrics was observed compared to the traditional logistic function. It was also noted that in binary classification tasks, SL can improve accuracy or sensitivity, depending on the sign of the asymmetry factor selected, predicting fewer false positives or fewer false negatives. It is concluded that the SL function offers a viable and promising alternative to conventional activation functions, providing better adaptation to asymmetric datasets.