Impact of global warming on sea level rise has become an important worldwide concern, challenging the coastal societies. Understanding and reducing uncertainties in sea level projections is a matter of increasing urgency seeking to provide guidance for assessing the impacts of the nearest future sea-level changes. The sea level projections are always accompanied by uncertainties, some of which are induced by the internal climate variability (ICV) originated from the complex climate system dynamics. Lack of attention to this irreducible uncertainty can lead to large under/overestimation of the future sea level rise impacts on coastal areas. From a decision strategy standpoint, this is a critical issue for the policy communities seeking to design adequate responses and to make adaptation strategy choices accounting for the impact of sea level rise in a warming world. Here, we estimate the ICV uncertainty in the future sea level rise by combining the large-ensemble experiments of the Community Earth System Model version 1 (CESM1-LE) with the power-law statistics. We evaluate, by 2100 and under RCP8.5, for the coastal mega-cities worldwide, the contribution of ICV to sea level rise and discuss the impact of this uncertainty in terms of coastal flooding. Thus, we show that if the ICV uncertainty upper limit is reached, certain regions of the world will be further jeopardized, as in Asia and tropical Pacific.