Risk is usually expressed as a combination of likelihood and consequence but obtaining credible likelihood estimates is difficult. The Conflicting Incentives Risk Analysis (CIRA) method uses an alternative notion of risk. In CIRA, risk is modeled in terms of conflicting incentives between the risk owner and other stakeholders in regards to the execution of actions. However, very little has been published regarding how CIRA performs in non-trivial settings. This paper addresses this issue by applying CIRA to an Identity Management System (IdMS) similar to the eGovernment IdMS of Norway. To reduce sensitivity and confidentiality issues the study uses the Case Study Role Play (CSRP) method. In CSRP, data is collected from the individuals playing the role of fictitious characters rather than from an operational setting. The study highlights several risk issues and has helped in identifying areas where CIRA can be improved.

Abstract. Today, with the advancement of information technology, there is a growing risk to privacy as identity information is being used widely. This paper discusses some of the key issues related to the use of game theory in privacy risk analysis. Using game theory, risk analysis can be based on preferences or values of benefit which the subjects can provide rather than subjective probability. In addition, it can also be used in settings where no actuarial data is available. This may increase the quality and appropriateness of the overall risk analysis process. A simple privacy scenario between a user and an online bookstore is presented to provide an initial understanding of the concept.

No abstract

Abstract. In a typical classical risk assessment approach, the probabilities are usually guessed and not much guidance is provided on how to get the probabilities right. When coming up with probabilities, people are generally not well calibrated. History may not always be a very good teacher. Hence, in this paper, we explain how game theory can be integrated into classical risk management. Game theory puts emphasis on collecting representative data on how stakeholders assess the values of the outcomes of incidents rather than collecting the likelihood or probability of incident scenarios for future events that may not be stochastic. We describe how it can be mapped and utilized for risk management by relating a game theoretically inspired risk management process to ISO/IEC 27005. This shows how all the steps of classical risk management can be mapped to steps in the game theoretical model, however, some of the game theoretical steps at best have a very limited existence in ISO/IEC 27005.

No abstract