The two major themes in the current construction industry are digital construction and low environmental impact. As a prominent digital construction technology, concrete 3D printing has attracted increasing attention. However, the current understanding of the durability of 3D printed cement-based materials (3DPCM) remains limited, which hinders its wider application, especially as load-bearing, reinforced concrete structures. This work shares the knowledge acquired during a broad interlaboratory study regarding the durability of 3DPCM with 15 laboratories from 13 countries participating, under the framework of TC 304-ADC ‘Assessment of Additively Manufactured Concrete Materials and Structures’. Anisotropy in water absorption capacity, carbonation and chloride ingress resistance of 3DPCM were evaluated by 15 institutes with their own printable materials and printing equipment. Additionally, the impacts of cold joints on these properties were investigated and a comparison between printed and cast samples was carried out. The outcome of this study indicates that the water absorption test provides information on the bulk porosity of the samples, while the carbonation and chloride ingress tests are more effective and visually reflect the local defects, especially the layer interfaces and cold joints. The water ingress depth of cast samples prepared with printable mixtures is an order of magnitude higher compared to conventional concrete, while their carbonation and chloride ingress resistance are comparable. The sorptivity and estimated water ingress height of printed samples measured in the direction parallel to the filaments is generally higher than that measured in the perpendicular direction and in cast samples. Similarly, the carbonation and chloride ingress depth and rate of printed samples measured in the direction parallel to the filaments is generally higher than that measured in the perpendicular direction or in cast samples. The overall durability of 3DPCM is weakened by anisotropy, these effects can be addressed with targeted mixture design and processing strategies. Due to the variations in printers, printing parameters and materials, three types of cross-section geometries were observed in printed samples with cold joints. The carbonation depth that measured from the maximum carbonation ingress point near the cold joint to the sample edge effectively captures the effect of cold joints in all these three types of cross-section geometries of printed samples. Finally, the participants identified areas of improvement in the methodology and suggestions were made to refine the procedure for adoption in future research.