A First Look into DeFi Oracles

TL;DR

This paper investigates the design and deployment of DeFi oracles, revealing their trust assumptions, operational issues, and deviations from actual exchange rates, and proposes potential improvements.

Contribution

It provides the first empirical analysis of DeFi oracles, highlighting trust issues, operational challenges, and deviations, along with suggestions for enhancements.

Findings

Oracles are often trusted with low accountability.

Prices from oracles frequently deviate from actual exchange rates.

Operational issues and anomalies are common in deployed oracles.

Abstract

Recently emerging Decentralized Finance (DeFi) takes the promise of cryptocurrencies a step further, leveraging their decentralized networks to transform traditional financial products into trustless and transparent protocols that run without intermediaries. However, these protocols often require critical external information, like currency or commodity exchange rates, and in this respect they rely on special oracle nodes. In this paper, we present the first study of DeFi oracles deployed in practice. First, we investigate designs of mainstream DeFi platforms that rely on data from oracles. We find that these designs, surprisingly, position oracles as trusted parties with no or low accountability. Then, we present results of large-scale measurements of deployed oracles. We find and report that prices reported by oracles regularly deviate from current exchange rates, oracles are not free from operational issues, and their reports include anomalies. Finally, we compare the oracle designs and propose potential improvements.