This Article explores the ability of technology—specifically, online judicial procedures—to eliminate systematic group-based litigation outcome disparities (i.e., disparities correlated with the visible identity markers of litigants). Our judicial system has long operated under the assumption that it can only be “impartial enough.” After all, judges, like all human beings, harbor implicit biases that often are sizable, unconscious, and triggered automatically, and research indicates that strategies to curb implicit biases in human decision-making may be ineffective, especially in the face of the resource and caseload constraints of modern-day adjudication. The recent emergence of online court proceedings, however, offers new hope for curtailing disparities. By allowing hearings to occur without face-to-face interactions, online legal proceedings may lessen the salience of group-identity traits, thereby mitigating unwarranted disparities and enhancing the impartiality of the justice system. Yet online proceedings differ from in-person hearings in ways beyond merely reducing the salience of age, gender, and race—and these differences may also influence group-based outcome disparities. Using state-court data, we study group-based disparities in online and offline civil infraction cases. We present evidence that is consistent with the existence of implicit or other structural biases in face-to-face proceedings; all else equal, legal outcomes appear to vary by litigant age and race but not by gender. These disparities fade with the change in medium, possibly by circumventing implicit biases. We explore the implications of our findings for system impartiality and weigh the challenges confronting efforts to realize substantive, procedural, and digital justice in online courts.