Predictive modeling is essential to better understanding and optimization of machining processes. Modeling of cutting forces has always been one of the main problems in metal cutting theory. In this paper, artificial neural networks (ANNs) were used for modeling correlations between cutting parameters and cutting force components in turning AISI 1043 steel. Cutting force components were predicted by changing cutting speed, feed rate, depth of cut and cutting edge angle under dry conditions. In order to improve generalization capabilities of the ANN models, Bayesian regularization is used in ANN training. Considering experimental data for ANN training, five ANN models were tested. For evaluating the predictive performance of ANN models, three performance criteria were given consideration. The overall mean absolute percentage error for cutting force components was around 3 %. This study concludes that Bayesian regularized ANN of quite basic architecture using small training data is capable of modeling multiple outputs with high prediction accuracy.