This study presents the development of models for predicting the slump and compressive strength of Alkali-Activated Fly Ash-Ground Granulated Blast Furnace Slag (AAFS) concrete, utilizing genetic algorithm (GA) and artificial neural network (ANN) techniques. The models are based on four input parameters, include %Na2O, %GGBS, W/S, and tpaste, and a total of 178 experiments were conducted to determine the slump and compressive strength of AAFS concrete. GA was used to optimize the hyperparameters of the ANN models, and to build the Mathematic model. The ANN models achieved R-squared values of 0.93 and 0.97 for slump and compressive strength, respectively, while the Mathematic model obtained R-squared values of 0.88 and 0.95 for slump and compressive strength, respectively. The results indicate that the compressive strength of AAFS concrete is significantly affected by %GGBS, %Na2O, and W/S, while the slump is most influenced by tpaste and W/S. Furthermore, each tpaste value corresponds to two threshold values of W/S, with a slight change in W/S between the two thresholds resulting in a significant change in the slump. Similarly, each W/S value corresponds to two threshold values of tpaste, where a low tpaste value initially leads to a very low slump. However, beyond a certain threshold, the slump increases rapidly until it reaches an upper threshold value.