AIM: This clinical retrospective study aimed to compare the deviations in single-tooth implant placement using novel semi-autonomous robotic-assisted surgery system (sa-RASS) and dynamic navigation system (DNS) methods. MATERIALS AND METHODS: A retrospective analysis of medical records from January to December 2023 was conducted to compare the implantation accuracy of the sa-RASS and DNS in partially edentulous patients using cone-beam computed tomography. Platform, apex, and angular deviations were measured and analyzed. The Kolmogorov-Smirnov test was used to check the data distribution, along with t-tests or Mann-Whitney U-tests, where appropriate. RESULTS: Fifty-seven patients (57 implants) were analyzed: 29 (29 implants) in the sa-RASS group and 28 (28 implants) in the DNS group. The comparison of platform, apex, and angular deviation between the sa-RASS group and DNS group were 0.91 ± 0.46 mm vs. 1.26 ± 0.51 mm (p <
0.001), 1.06 ± 0.60 mm vs. 1.51 ± 0.56 mm (p <
0.001), and 3.07° ± 1.66° vs. 3.71° ± 1.64° (p >
0.05), respectively. In addition, there was no significant difference in the accuracy of different implant regions (premolar, molar, maxilla, and mandible) or implant length (p >
0.05). CONCLUSIONS: In the present study, sa-RASS implant placement showed better positional accuracy than DNS implant placement in platform and apex deviation, although these improvements in accuracy may have limited clinical relevance, suggesting that the sa-RASS might be advantageous in dental implant surgery. TRIAL REGISTRATION: ClinicalTrials.gov identifier: ChiCTR2400085089.