The increasing use of Adhesively-bonded joints in industrial applications resulted in more attention to damage assessment in these joints. The aim of the present study is to characterize the damage in bi-material double-lap adhesively-bonded joints by Acoustic Emission (A E). Two different structural adhesives, representing a ductile (Methacrylate-based) and brittle (epoxy-based) types, were used to bond C F R P skins to a steel core. The fabricated joints were loaded in tension while damage evolution was monitored by A E . Due to the difference in the fracture nature of the adhesives "brittle vs. ductile", different damage mechanisms occurred in the specimens; including adhesive layer failure, steel deformation, adhesive/adherends interfacial debonding and delamination in the C F R P skin. In order to distinguish and classify these damage types by A E, the AE features of each damage mechanism were first obtained by conducting some standard tests on the individual constituent materials. Then, these AE reference patterns were used to train an ensemble decision tree classifier. Finally, the trained model classified the AE signals of the double-lap tests and the images captured by camera were utilized to verify AE results. This study demonstrates the potential of AE technique for damage characterization of the adhesively-bonded bi-material joints.