Robotic-assisted total knee arthroplasty (RATKA) is an alternative surgical treatment method to conventional total knee arthroplasty (COTKA) that may deliver better surgical accuracy. However, its impact on patient outcomes is uncertain. The aim of this systematic review of randomized controlled trials (RCTs) is to evaluate whether RATKA could improve functional and radiological outcomes compared with COTKA in adult patients with primary osteoarthritis of the knee. We searched Ovid MEDLINE, EMBASE, Scopus, and the Cochrane Library to identify published RCTs comparing RATKA with COTKA. 2 reviewers independently screened eligible studies, reviewed the full texts, assessed risk of bias using the Risk of Bias 2.0 tool, and extracted data. Outcomes were patient-reported outcomes, range of motion, and mechanical alignment (MA) deviation and outliers, and complications. We included 12 RCTs involving 2,200 patients. RATKA probably results in little to no effect on patient-reported outcomes (mean difference (MD) in the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) score of -0.35 (95% confidence interval [CI] -0.78 to 0.07) and range of motion (MD -0.73°; CI -7.5° to 6.0°) compared with COTKA. However, RATKA likely results in a lower degree of MA outliers (risk ratio 0.43; CI 0.27 to 0.67) and less deviation from neutral MA (MD -0.94°; CI -1.1° to -0.73°). There were no differences in revision rate or major adverse effects associated with RATKA. Although RATKA likely results in higher radiologic accuracy than COTKA, this may not be clinically meaningful. Also, there is probably no clinically important difference in clinical outcomes between RATKA and COTKA, while it is as yet inconclusive regarding the revision and complication rates due to insufficient evidence.