Ultra-deep carbonate reservoirs in Southwest China are characterized by deep burial (6000–8000 m), high formation temperature (140–180 °C), and strong non-homogeneity. The acid filtration loss during acid fracturing reforming is significant, the reaction rate of acid rock is fast, and the length of acid-etched fracture is limited. These factors result in a significant difference in the effect of acid fracturing on raising production. The effectiveness of acid fracturing on deep carbonate reservoirs is evident in the production rate of 25% of the wells, which can reach more than 10 × 104 m3/d. However, 25% of the wells do not show any production rate after acid fracturing. Therefore, it is crucial to conduct a study on the main controlling factors of the acid fracturing effect and design a more targeted acid fracturing scheme and process. This paper presents an artificial intelligence-based approach to evaluate the primary controlling factors of acid fracturing effects in deep carbonate rocks. The specific steps involve preprocessing the data to improve its quality, followed by a comprehensive analysis of the factors influencing the effect of acid fracturing using the Random Forest and Recursive Elimination Algorithm. Finally, a production prediction model is established using the neural network algorithm. The combination of influencing factors with the best prediction effect is determined as the final main controlling factor. This will provide support for the subsequent optimization of the acid fracturing process parameters in deep carbonate rocks, enhancing production and efficiency. The study results indicate that selecting six influencing factors for prediction yields the best effect for this block. At this point, the mean square error is 0.02 and the correlation coefficient is 0.9, demonstrating the method's applicability to carbonate reservoirs and high prediction accuracy. The main controlling factors affecting the effect of acidic fracturing in this block are the total length of the stimulation section, the storage coefficient, the number of segments, the acid fracturing process, and the rate of pressure drop. This study presents a new understanding of the main controlling factors that affect the effect of acid fracturing on ultra-deep carbonate rocks, based on the proposed evaluation method. The findings will provide support for optimizing the acid fracturing process parameters.