Purpose: This study evaluates the impact of metal artifacts in cone-beam computed tomography (CBCT) on the accuracy of static and dynamic computer-assisted implant surgery (CAIS) techniques. Materials and Methods: An implant was placed on each of thirty 3D-printed models embedded with Cobalt-Chrome strips to simulate metal artifacts by Porcelain-fused-to-metal (PFM), utilizing these CAIS techniques: radiographic template (RT) (n=10), radiographic markers (RM) (n=10), and dynamic navigation (DN) (n=10). Trueness and precision were analyzed by comparing 3D global deviation and the difference in implant positions at the neck, apex, depth, and angle in initial planned and final placed scans. Results: DN exhibited significantly lower 3D global trueness deviation to RT (p = 0.022) and lower angular deviation compared to both RT (p = 0.003) and RM (p = 0.002). RM showed greater trueness at the implant neck compared to DN (p = 0.005) and better depth trueness than RT (p = 0.001) and DN (p = 0.027). DN demonstrated higher precision in implant angulation than RM (p = 0.011) and RT (p = 0.041). Conclusion: In the presence of metal artifacts, both DN and RM techniques offered greater trueness in specific implant positions compared to RT. However, DN proved to be the most precise method for implant placement.