SUMMARY Objectives: To evaluate and compare the fracture toughness of dental CAD/CAM materials of different material classes intended for in-office milling (glass ceramics, hybrid, resin composites) and the influence of aging on this property. Methods and Materials: The fracture toughness (critical intensity factor, K Ic ) values of 9 CAD/CAM restorative materials (Ambarino High-Class, Brilliant Crios, Cerasmart, exp. CAD/CAM composite, Katana Avencia, Lava Ultimate, VITA Enamic, IPS Empress CAD, and IPS e.max CAD) were determined using the SEVNB method in a four-point bending setup. Twenty bending bars of each material with a 4 × 3 cross and a minimum length of 12 mm were cut out of CAD/CAM milling blocks. Notching was done starting with a pre-cut and consecutive polishing and v-shaping with a razor blade, resulting in a final depth of v-shaped notches of between 0.8 and 1.2 mm. Half of the specimens were selected for initial fracture toughness measurements. The others were thermocycled in distilled water for 30,000× (5/55°C; 30-second dwell time) before testing. Specimen fracture surfaces were analyzed using confocal laser scanning microscopy. Results: All specimens for each material fractured into two fragments and showed the typical compression curl and brittle failure markings. Comparing initial K Ic values, lithium disilicate ceramic IPS e.max CAD showed significantly the highest and leucite-reinforced ceramic IPS Empress CAD significantly the lowest K Ic values ( p <0.001). All tested CAD/CAM materials with a resin component ranged in the same K Ic value group ( p >0.999-0.060). After thermal cycling, the highest K Ic values were measured for lithium disilicate ceramic IPS e.max CAD, followed by resin composite materials Ambarino High-Class ( p <0.001-0.006) and hybrid material VITA Enamic ( p <0.001-0.016), while the significantly lowest values were reflected for the resin composite materials Cerasmart, LAVA Ultimate ( p <0.001-0.006), and Katana Avencia ( p <0.001-0.009). The roughness of the fracture surfaces varied depending on the microstructure of the respective material. The ceramic surfaces showed the smoothest surfaces. The fracture surface of VITA Enamic revealed microstructural inhomogeneities and microcracks. For CAD/CAM resin composite materials, crack paths through the matrix and interfaces of matrix and fillers could be observed at the microstructure level. Conclusions: The materials tested show differences in fracture toughness typical for the class they belong to. With one exception (Ambarino High-Class), thermocycling affected the fracture toughness of materials with a resin component negatively, whereas the leucite and lithium disilicate ceramic showed stability.