Upper extremities limitation is a common functional impairment in patients with Multiple Sclerosis (PwMS). Novel technological devices are increasingly used in neurorehabilitation to support motor function improvement and the quantitative assessment of motor performance during training in patients with neurological diseases. In this review, we systematically report the evidence on clinical applications and robotic-assisted arm training (RAT) in functional recovery in PwMS. PubMed/MEDLINE, the Cochrane Library, and the Physiotherapy Evidence Database (PEDro) databases were systematically searched from inception to March 2021. The 10-item PEDro scale assessed the study quality for the RCT, and the AMSTAR-2 was used to assess the quality of the systematic review. The 5-item Oxford CEBM scale was used to rate the level of evidence. A total of 10 studies (161 subjects) were included. The selected studies included one systematic review, four RCTs, one randomized crossover, and four case series. The RCTs were scored as high-quality studies, while the systematic review was determined to be of low quality. Shoulder range of motion, handgrip strength, and proximal arm impairment improved after RAT. Manual dexterity, arm function, and use in daily life also ameliorated arm function. The high clinical heterogeneity of treatment programs and the variety of robot devices affects the generalizability of the study results; therefore, we emphasize the need to standardize the intervention type in future studies that evaluate the role of robotic-assisted training in PwMS. Robot-assisted treatment seems safe and useful to increase manual dexterity and the quality of movement execution in PwMS with moderate to severe disability. Additional studies with an adequate sample size and methodological rigour are warranted to drive definite conclusions.