This study compared the adhesion of metal brackets bonded to different CAD/CAM materials after various surface conditioning methods. CAD/CAM blocks (N=204, n=17 per group) of (a) VITA Mark II (VM), (b) IPS e.max CAD (IP), (c) Lava Ultimate (LU), and (d) VITA ENAMIC (VE) were conditioned with one of the following methods: C-Control: (fine diamond bur); CJ: (fine diamond bur+air-abrasion with 30 mu m SiO2+silane), and HF: (fine diamond bur +9.5% hydrofluoric acid+silane). Metal brackets were bonded to the conditioned surfaces of the specimens, stored in artificial saliva for 24h at 37 degrees C and thermocycled (x1000). Subsequently, the brackets were debonded under shear in a Universal Testing Machine (1mm/min). Failure types were analyzed under scanning electron microscope. Data were analyzed using two-way ANOVA and Tukey's tests (alpha=0.05). Two-parameter Weibull distribution values, including the Weibull modulus, scale, and shape, were calculated. Mean bond strength (MPa) values were significantly affected by the surface conditioning method (p<0.001) but not the CAD/CAM material type (p=0.052). Bond strengths for all CJ and HF-conditioned specimens were twofold higher (11.83 +/- 1.95 - 9.44 +/- 1.63) than those for control specimens with all materials (4.73 +/- 0.93 - 6.02 +/- 0.69). Lower mean values were obtained in LU-CJ (9.78 +/- 1.61) and LU-HF (9.44 +/- 1.63) than those for other groups (11.83 +/- 1.95 - 10.93 +/- 1.33) groups (p<0.05). Weibull distribution showed higher shape values for VM-CJ (11.26) and VM-HF (8.87) than those for other groups (0.82-1.83). In control groups, exclusively adhesive and after conditioning mainly mixed failures were observed. Chairside silica coating or HF conditioning significantly improved metal bracket adhesion to both glassy matrix and hybrid CAD/CAM materials tested, with the most reliable adhesion being observed for feldspathic ceramic.