Fifteen years have elapsed since a report was released by the National Academy of Sciences (NAS) on the scientific status of polygraph testing. The NAS report concluded that the scientific basis of the comparison question technique (CQT) was weak, the extant research was of low quality, the polygraph profession's claims for the high accuracy of the CQT were unfounded, and, although the CQT has greater than chance accuracy, its error rate is unknown. Polygraph proponents argue that current research indicates that the CQT has 90% or better accuracy, the National Research Council of the National Academy of Sciences' (2003) analysis supports this accuracy claim, and the CQT qualifies as legally admissible scientific evidence. We review the scientific literature that has appeared since the appearance of the NAS publication, including a new method for estimating polygraph accuracy. We show that the NAS report has been misrepresented and misinterpreted by those who support use of the CQT in forensic settings. We conclude that the quality of research has changed little in the years elapsing since the release of the NAS report, and that the report's landmark conclusions still stand. Public Significance Statement Lie detection tests have profound life-altering consequences ranging from public perceptions of an examinee's credibility and integrity to effects on job security and possible incarceration. In 2003, the National Academy of Sciences concluded that polygraph testing had a weak scientific basis and unknown error rate. Analysis of research conducted over the last 15 years indicates that these conclusions remain valid. As polygraph tests are routinely used by law enforcement agencies in North America and other countries, it is vital that the public be aware of their shortcomings.