Cs 3 Cu 2 I 5 and CsCu 2 I 3 have shown potential applications as optoelectronic materials, but the mechanism of their emission remains controversial. It is widely believed that the Stokes shifts of Cs 3 Cu 2 I 5 (1.45 eV) and CsCu 2 I 3 (1.63 eV) are caused by self-trapped excitons, but the apparent changes in emission intensity observed experimentally are difficult to explain by this theory. In this paper, the defect emission mechanism is investigated: Iodine vacancies (V I ) are widely present in iodide materials, but the effect of such point defects on Cs 3 Cu 2 I 5 and CsCu 2 I 3 has been little studied. Theoretical calculations based on the first-principles suggest that the V I defect (3.05 eV) may be directly related to the Stokes shift of Cs 3 Cu 2 I 5 (similar conclusions were obtained for CsCu 2 I 3 ). In the experiments related to V I , the emission quenching in the iodination treatment is associated with the decrease of defect concentration, while the emission recovery in the annealing treatment stems from the increase of defect concentration caused by iodine decomposition. The iodination and annealing treatments for CuI shows that V I defects play an important role in the emission of iodide. In addition, experiments to control the defect concentration by RTP treatment also showed that V I defects have an effect on the emission intensity of Cs 3 Cu 2 I 5 and CsCu 2 I 3 . Based on the experimental results and theoretical calculations, the emission mechanism of Cs 3 Cu 2 I 5 and CsCu 2 I 3 may be the semiconductor defect emission theory. We believe that our research on the defect emission theory will provide new insights into the application of Cs 3 Cu 2 I 5 and CsCu 2 I 3 . • The emissions of Cs 3 Cu 2 I 5 and CsCu 2 I 3 were analyzed by semiconductor defect emission theory. • The effect of V I defects on Cs 3 Cu 2 I 5 and CsCu 2 I 3 was investigated by theoretical calculations. • Iodination and annealing treatments were designed to support the defect emission theory. • Similar treatments were performed for other iodides (CuI) to demonstrate that the emission was associated with V I defects. • RTP treatment was performed on CuI to control V I defect concentration in Cs 3 Cu 2 I 5 and CsCu 2 I 3 .