The Louis Bachelier Institute has just released the fifth issue of its “ILB Methods” collection, devoted to the risk of cyber claims accumulation for insurance. It was produced with the participation of Caroline Hillairet, professor of actuarial sciences at Ensae-Paris (Crest) and Olivier Lopez, professor of applied mathematics at Sorbonne University

Cyber risks have multiplied in recent years, particularly against companies. This trend was notably marked in May 2017 by the Wannacry ransomware attack on more than 200,000 victims worldwide (companies, individuals, health services). And during the Covid-19 pandemic, cyber risks even increased. The recent attacks against Colonial Pipeline, an American oil pipeline operator, which had to cease its activities for several days, and against an Asian subsidiary of the insurer Axa, bear witness to this.

Cyber contagion modelling similar to epidemiology

To deal with these attacks, which can lead to considerable losses for certain entities that are little or unprepared, especially SMEs, insurance products are developing strongly (prevention, compensation, etc.). However, in the event of a global attack, such as Wannacry, the dynamics of contagion to hundreds of thousands of victims may be difficult to estimate and constitute a risk of accumulation of claims for insurance companies.

In order to model the contagious dynamics of a cyber attack at the global level, Caroline Hillairet and Olivier Lopez propose to draw inspiration from the compartmental models used in epidemiology.

For a detailed explanation of their methodologies, see the ILB Methods below: