PRA (Sweeney, 2019), ACPR (ACPR, 2019) and EIOPA (EIOPA, 2020), among others, have highlighted the importance of silent cyber risk assessment. We present in this paper a toy portfolio inspired by OECD sectors data where the interactions between the insured are modeled through a network. We first define the portfolio’s silent exposure. Then, using a business interruption scenario, we evaluate the number of infected nodes and the probable loss using epidemiological models. From this, we see that adjustments to certain parameters (insurer’s intervention capacity or the sectors in the portfolio) could reduce the overall global loss. The peaks of contamination over time are also illustrated for each sector.