Abstract: The present disclosure provides an artificial intelligence detection system for deep-buried fuel gas pipeline leakage, including a multi-field source information collecting system, a data processing and analyzing system, and a monitoring and warning system, wherein the multi-field source information collecting system includes a concentration field collecting subsystem, a temperature field collecting subsystem, and a geoelectric field collecting subsystem; the concentration field collecting subsystem collects concentration field data; the temperature field collecting subsystem collects temperature field data; the geoelectric field collecting subsystem collects geoelectric field data; the data processing and analyzing system receives the concentration field data, temperature field data and geoelectric field data, calculates variations of the respective data, compares the variations with corresponding variation thresholds, and determines whether to generate a warning signal; the monitoring and warning system ala