When good blocks go bad: Managing unwanted blockchain data

Abstract Blockchain has been praised for providing the technical infrastructure that enables a group of self-interested entities to share data without relying on intermediaries. Technically, blockchain is a distributed and decentralized append-only database. This latter aspect leads to an important, yet overlooked governance issue, namely what should the network members do when erroneous or malicious data are added to the blockchain ledger? We start by describing three public cases when the above happened. For each case, we elaborate on the adopted solution, which we refer to as the “rollback,” the “do nothing,” and the “overturn” solution. Drawing from these previous cases, discussions with experts, and from our own experience with blockchain research and development, we provide suggestions concerning managerial, technical, and information security policies and practices organizations should follow when contemplating enterprise-level applications of the blockchain technology.