For 30 years, bright light therapy (BLT) has been considered as an effective, well-tolerated treatment for seasonal affective disorder (SAD). Because of low response rates, new treatment strategies are needed for bipolar depression (BD), which resembles SAD in certain respects. Few placebo-controlled studies of BLT efficacy have been carried out for BD. Accordingly, this study evaluates the efficacy and safety of BLT as an add-on treatment for BD. Thirty-two BD outpatients were randomly assigned to BLT (10000 lx) or dim light (DL, < 500 lx). During a two-week period, light was administered each morning for 30 min. The Hamilton Rating Scale for Depression and the Montgomery-Asberg Depression Rating Scale assessed clinical outcome, and the UKU Side Effects Rating Scale evaluated side effects. No significant difference was observed in baseline depression scores in the two groups. Response rates for BLT and DL were 81% and 19%, and remission rates were 44% and 12.5%, respectively. Analyses showed statistically significant reductions in depression scores for the BLT group compared with the DL group on all scales. Side effects were similar in both groups, with headache as the most common side effect. The results suggest that BLT is an effective and safe add-on treatment for BD.