Computer‐aided psychotherapy (CP) is said to (1) be as effective as face‐to‐face psychotherapy, while requiring less therapist time, for anxiety disorder sufferers, (2) speed access to care, and (3) save traveling time. CP may be delivered on stand‐alone or Internet‐linked computers, palmtop computers, phone‐interactive voice response, DVDs, and cell phones. The authors performed a meta‐analysis of 23 randomised controlled studies (RCTs) that compared CP with non‐CP in anxiety disorders: phobias, n = 10; panic disorder/agoraphobia, n = 9; PTSD, n = 3; obsessive–compulsive disorder, n = 1. Overall mean effect size of CP compared with non‐CP was 1.08 (95% confidence interval: 0.84–1.32). CP and face‐to‐face psychotherapy did not differ significantly from each other (13 comparisons, d = −0.06). Much caution is needed when interpreting the findings indicating that outcome was unrelated to type of disorder, type of comparison group, mode of CP delivery (Internet, stand‐alone PC, palmtop), and recency of the CP system and that effect size decreased when more therapist time was replaced by the computer. Because CP as a whole was as effective as face‐to‐face psychotherapy, certain forms of CP deserve to be integrated into routine practice.