Investigating performance-based incentive mechanisms in Chainlink

Abstract

Blockchain oracles serve as the link between real-world data and blockchain ecosystems. This external data plays a pivotal role in decentralised applications, such as DeFi protocols, which manage substantial amounts of users’ assets. Consequently, oracle operators bear huge responsibilities because inaccuracies in data delivery can result in severe consequences for protocols and the entire blockchain in general. Furthermore, given the critical nature of this data, oracles are susceptible to various attacks aimed at manipulating the information to benefit attackers, which result in losses amounting to hundreds of millions of U.S. dollars. Furthermore, despite the recent increase in academic studies related to blockchain oracles, there still seems to be a gap when it comes to incentive mechanisms. Additionally, beyond shouldering significant responsibilities and facing potential damage to their reputation, oracle operators also bear substantial costs when delivering external data onto the blockchain. These costs include labour costs for man-power, hardware resources and data acquisition expenses from various sources. Consequently, it becomes imperative to establish robust incentives for operators, not only as compensation for their efforts but also as a deterrent against potential misconduct that could result in substantial losses. Despite not being a straightforward task due to several factors and issues outlined in prior studies, it is logical to combine incentivise operators based on their performances. This strategy not only encourages operators to provide more accurate data but also elevates the overall quality of the final oracle solution data. In light of this, this study delves into the realm of performance-based incentivisation mechanisms within Chainlink’s oracle network. The study’s core objectives were to assess the existing Chainlink incentive system’s alignment with operators’ performances and to devise improvements for this mechanism. The investigation unveiled that while the current scheme predominantly promotes performance by refraining from incentivising missed submissions, it does not use submissions’ accuracies as basis for incentivisation. Subsequently, this study introduced a performancebased incentive mechanism and a penalisation scheme. This incentive mechanism with the penalisation scheme was also tested on another Chainlink feed, as a validation data set, to evaluate the relationship between incentives and performances. Finally, this study sought feedback from some Chainlink operators to perform a qualitative analysis of their opinions regarding performance-based incentivisation mechanisms in Chainlink.