RiskLayer Proposes Multi-Agent Influence Diagrams for DeFi Governance

• RiskLayer’s new paper introduces Multi-Agent Influence Diagrams to enhance DeFi governance and tackle economic vulnerabilities.
• Recent governance attacks highlight the need for secure models, with MAIDs offering solutions to align incentives within protocols.
• The RiskLayer team plans to implement MAIDs on Compound Finance, aiming to use open-source tools to strengthen DeFi governance.

Sudipan from RiskLayer has announced a development in decentralized finance (DeFi) governance. The team at Chain Risk has released their second version of the paper on Multi-Agent Equilibrium in Governance Models. This ambitious research aims to tackle Governance Attacks and Governance Economic Vulnerabilities (GEV) for a more equitable crypto future.

The paper delves into the complexities of DeFi governance. Multiple independent agents with differing incentives often complicate decision-making. The authors introduce Multi-Agent Influence Diagrams (MAIDs) as a crucial analytical tool to address this. MAIDs provide a comprehensive framework for understanding how agents interact and influence governance outcomes.

Understanding MAIDs in Governance

MAIDs model strategic interactions within DeFi systems. The research focuses on a simplified governance game that mimics real governance protocols. By computing Nash equilibria through MAIDs, the authors analyze how local decisions affect overall governance. This modelling technique also considers market trends and external factors impacting governance choices.

Additionally, the paper explores MAIDs’ application within MakerDAO. MAIDs extend traditional frameworks to account for the intricate dynamics between multiple decision-makers. They incorporate various variables, such as preferences and actions, allowing a thorough examination of cooperation and negotiation.

Addressing Recent Governance Attacks

Following the recent governance attack on Compound Finance, the need for economic security around governance has become crucial. A whale, known as Humpy, manipulated $COMP token delegations to influence votes. 

Humpy aimed to divert $24 million in COMP to a protocol he controlled, named goldCOMP. This situation highlights how governance tokens can be exploited by a few, jeopardizing the protocol’s integrity.

In response, MAIDs offer a promising solution. They can model and mitigate vulnerabilities by formally verifying incentive structures within governance systems. The research emphasizes the importance of aligning agents’ goals with the protocol’s objectives to enhance sustainability and security .

The RiskLayer team plans to model MAIDs with Proof of Concept on Compound Finance. They aim to derive experimental results and publish a formal paper for the industry. Ultimately, they envision creating an open-source MAID implementation across all major governance structures, enhancing the resilience of DeFi governance.