Abstract: The corrosion problem in the conventional steel reinforcement in concrete structures has diverted the researchers to explore alternative materials. As a substitute to replace the traditional steel bars in reinforced concrete structures, innovative reinforcement such as fiber reinforced polymer (FRP) rebars has been suggested. Consequently, different codes and guidelines have been proposed for forecasting the shear strength of FRP reinforced members using traditional empirical methods and neural networks. The current paper concentrates on the development of a hybrid intelligence model, namely an artificial neural network articulated with a Bayesian optimization algorithm (ANN-BOA) for estimating the shear strength of these types of members without stirrups. Totally, 216 specimens, collected from the literature, were used in this analysis. The input parameters of the model were beam depth, the ratio of shear span and depth, effective reinforcement ratio, and concrete strength. The ANN hyperparameters (viz. neuron numbers in the hidden layer, learning rate) have been tuned automatically to get the best predictions. The estimated shear strengths are compared with the recent design provisions of Japan (JSCE), UK (BISE), Italy (CNR-DT 203), Canada (CHBDC, CSA S806), and the USA (ACI 440.1R). The results were also compared with a similar ANN model. It was observed that the predicted results using the proposed method are better than those of the other methods in terms of some statistical as well as performance measuring parameters with a maximum Pearson correlation coefficient (R) value of 0.97. These values are higher than the other investigated methods.