The Royal Canadian Mounted Police implemented the Phased Interview Model in Canada and has argued that it is a novel and productive way to interview suspects. We applaud the Royal Canadian Mounted Police for moving away from a purely accusatorial approach and recognize that Phased Interview Model contains several science-based practices. In this article, however, we evaluate the Phased Interview Model critically. In particular, we present compelling empirical evidence that three fundamental practices (minimizing culpability, mischaracterizing evidence, and asking leading questions) in the Phased Interview Model put the truth-seeking function of police interviews at risk. We also explore the challenges inherent in combining accusatorial and information gathering techniques into a hybrid ‘toolbox’ approach. We conclude that advocating for interview protocols that contain dangerous or untested practices may hinder the Royal Canadian Mounted Police’s ability to achieve their purported goals of obtaining voluntary statements and accurate information.