Current understanding of the pathophysiology and epidemiology of atrial fibrillation (AF) is based primarily on studies in White populations of European ancestry with limited data on the non-White populations. Many traditional and novel risk factors for AF have been already established. However, little is known about the differential impact of these AF risk factors in non-White versus White populations. Few studies have compared the prevalence and incidence of AF in Black versus White individuals, and still fewer extended this comparison to other races and ethnicities. These few studies consistently reported that Black individuals have lower AF burden compared with their White counterparts. These findings are not consistent with the reported high stroke rates in Black populations and the strong association between stroke and AF. In this perspective we argue that differential AF underascertainment among different racial and ethnic groups may be the key in explaining this paradox of racial/ethnic distribution of AF. Results from recent publications suggest that Black individuals would have higher rates of AF if more sensitive AF detection methods are used. Although these findings do not reach the level of definitive evidence, they may support the need for rethinking the epidemiology of AF across races and ethnicities in the context of the limited methodology available to detect AF, as well as the questionable assumption that the ability to diagnose AF across different racial/ethnic groups is the same.