The identification of indirect markers of suicide that do not rely on explicit self-report of suicide ideation is vital, as research indicates that a significant number of individuals who die by suicide do not discuss or explicitly deny thoughts of suicide with a provider in their last health care interaction. The current study tests whether a new measure of implicit association with suicide, a modified version of the Affect Misattribution Procedure (AMP), predicts suicide ideation in participants oversampled for the experience of suicide ideation. Results indicated that implicit associations with suicide-related images were positively correlated with levels of suicide ideation and predicted suicide ideation when entered as a simultaneous predictor with symptoms of depression, thwarted belongingness, and perceived burdensomeness. Implicit associations with positive and negative images were unrelated to thoughts of suicide. The modified version of the AMP may be a particularly useful indirect measure of suicide ideation that can be easily included in suicide risk assessment paradigms in clinical practice and research studies.