Suppose an agent has positions of various assets. As time goes by, the corresponding total wealth varies. Since the agent would like to keep the positions of the different assests for various different time periods, the position process will not be adapted. A natural question is how one can measure the dynamic risk of such kind of position process? It turns out that the theory of backward stochastic Volterra integral equations (BSVIEs, for short) presents a suitable framework for dynamic risk measures. In this talk, we will briefly present the relevant results on BSVIEs and construct a class of dynamic risk measures.