The conventional methods used in the design of drilled shafts might not fully consider the multiple sources of uncertainty in the soil such as the geometric and mechanical variability and/or the construction methods. These uncertainties can introduce nonlinearity to the analysis, leading to underestimation or overestimation of the resistance, which can be translated into expensive or even unsafe projects. Because of this, machine learning techniques such as artificial neural networks (ANNs), proven to be effective in solving nonlinear problems, are becoming popular in solving civil engineering problems. Therefore, the objective of this preliminary study is to evaluate a concept for predicting the nominal side resistance of drilled shafts with improved accuracy, using ANN. In this study, 45 load tests were collected from the extended version of the Nevada Deep Foundation Load Test Database and divided into 85% for training and 15% for testing. Then 1,638 ANN models were trained to determine the optimum model with a root-mean-squared error of 2,058 kips and an R-squared ( R2) of 85% on unseen load tests. The model was then benchmarked against the AASHTO predicted side resistance, with an average overall improvement of the prediction accuracy using ANN of 23%. This paper demonstrates that such models could be developed, improved, and used in the industry at an early stage and with limited data as supplemental tools that can help optimize designs in regard to construction safety, time, and cost.