Abstract: The sequential detection method is commonly used for detecting air leakage pathways in gas drainage boreholes. This method involves sequentially measuring the gas concentration at various measuring points and identifying the leakage location through comparative analysis. However, for long boreholes or those with numerous measuring points, the on-site detection efficiency of the sequential detection method is low. To address this issue, this paper proposed a rapid detection method for air leakage pathways in gas drainage boreholes based on the bisection method. Multiple measuring points were arranged axially within the borehole. A probe rod and a gas drainage pipe were used to separate the gas at a specific measuring point from the mixed gas inside the borehole. First, the gas concentration at the borehole bottom measuring point was measured. If it was significantly higher than the average in-hole gas concentration, the presence of an air leakage pathway in the borehole was confirmed. Subsequently, the in-hole measuring points were bisected, and the gas concentration at the intermediate measuring point was measured. If this concentration was close to the average in-hole gas concentration, the air leakage pathway was determined to be located in the section from that point to the borehole bottom; if it was significantly higher, the pathway was determined to be in the section from that point to the borehole collar. The section identified as containing the air leakage pathway underwent further recursive bisection. By comparing the gas concentration at the intermediate measuring point within a section with the average in-hole gas concentration, the localization range was progressively narrowed, which ultimately achieved the precise and rapid localization of the air leakage pathway. The field experiment results showed that the proposed method accurately identified the location of the air leakage pathway. Compared with the sequential detection method, the detection time was shortened by 26.5%. This improvement demonstrates that the proposed method effectively improves detection efficiency.