Uncertainty plays an important role in clustering. For example in customer segmentation we may be faced with the situation that a certain customer not necessarily belongs to just one segment, i.e. his/her class assignment is uncertain. Several cluster algorithms have been proposed that employ uncertainty modeling in different ways. The most frequently used techniques are probability theory, fuzzy logic, and recently rough sets. If uncertainty modeling is already important in static clustering this becomes even more important in dynamic clustering where several elements of the respective cluster can change over time. Changes produce uncertainty and that is where uncertainty modeling in dynamic clustering comes into play. In this paper we present briefly two cluster algorithms that employ soft computing approaches and provide a comparison regarding their capabilities to capture uncertainties in dynamic environments. Future research issues for this area are also identified.