Incentives in Security Protocols

摘要

Real world protocols often involve human choices that depend on incentives, including when they fail. We look at three example systems (the EMV protocol, consensus in cryptocurrencies, and Tor), paying particular attention to the role that incentives play in fail-safe and fail-deadly situations. In this position paper we argue that incentives should explicitly be taken into account in the design of security protocols, and discuss general challenges in doing so.