The question of how oil supply news shocks affect the economy is a current topic of research in macroeconomics. To address this question, a new model called the Factor Bayesian Additive Regression Tree model has been developed. This model is a nonlinear factor-augmented vector autoregression model that can handle large datasets and allow for nonlinear relationships to emerge from the data. The model is applied to a large US macro-financial dataset that includes oil supply news shocks. The results show that negative oil supply news shocks have a stronger and more lasting impact on the economy than positive shocks of the same size. The impact of oil supply news shocks is particularly pronounced in industrial production, financial variables, and equity prices. The effect of these shocks on employment also varies significantly across different US states, with manufacturing-intensive regions experiencing stronger contractions. Energy-producing states, on the other hand, experience partially offsetting dynamics after negative oil supply news shocks. The relationship between oil supply news shocks and their impact on the economy is nonlinear, with small shocks having a weak effect and moderate shocks having a more significant impact. Overall, the study provides new insights into how oil supply news shocks transmit to the economy and highlights the importance of considering nonlinear relationships and regional differences.