Purpose To evaluate the accuracy of the remote-controlled robotic computed tomography (CT)-guided needle insertion in phantom and animal experiments. Materials and Methods In a phantom experiment, 18 robotic and manual insertions each were performed with 19-gauge needles by using CT fluoroscopic guidance for the evaluation of the equivalence of accuracy of insertion between the two groups with a 1.0-mm margin. Needle insertion time, CT fluoroscopy time, and radiation exposure were compared by using the Student t test. The animal experiments were approved by the institutional animal care and use committee. In the animal experiment, five robotic insertions each were attempted toward targets in the liver, kidneys, lungs, and hip muscle of three swine by using 19-gauge or 17-gauge needles and by using conventional CT guidance. The feasibility, safety, and accuracy of robotic insertion were evaluated. Results The mean accuracies of robotic and manual insertion in phantoms were 1.6 and 1.4 mm, respectively. The 95% confidence interval of the mean difference was -0.3 to 0.6 mm. There were no significant differences in needle insertion time, CT fluoroscopy time, or radiation exposure to the phantom between the two methods. Effective dose to the physician during robotic insertion was always 0 μSv, while that during manual insertion was 5.7 μSv on average (P < .001). Robotic insertion was feasible in the animals, with an overall mean accuracy of 3.2 mm and three minor procedure-related complications. Conclusion Robotic insertion exhibited equivalent accuracy as manual insertion in phantoms, without radiation exposure to the physician. It was also found to be accurate in an in vivo procedure in animals. © RSNA, 2017 Online supplemental material is available for this article.