Forecasting prices of energy commodities, particularly oil, is very intricate, mainly due to the nonlinearity and high volatility associated with financial markets. In this research, a novel application of artificial neural networks for predicting oil prices over a 2-month horizon is proposed. The model is developed based on a rich dataset that included data per day for 24 years, ranging from 2001-01-05 to 2024-08-25, of main global stock indices, oil, and gold, with 5430 records per index. This current study employed a feedforward backpropagation artificial neural network (ANN), specifically designed to capture complex, nonlinear relationships between these variables. The ANN model was rigorously trained and tested, demonstrating exceptional predictive accuracy. The findings highlight the model’s potential as a robust tool for energy market prediction, offering a significant advancement over traditional forecasting methods by effectively handling large-scale, nonlinear data.