Deception detection is a longstanding research goal of considerable interest to the military, law enforcement, corporate security, social services, and mental health workers. However, both humans and polygraphs are very poor at this task. In this webinar, we describe classifiers we have developed to detect deception automatically from speech which are significantly better than humans. We are currently extending our studies to identify cues in speech that is trusted vs. mistrusted by humans and developing methods which may be able to improve human performance as well.