Ethereum co-founder Vitalik Buterin has put forth a significant proposal to overhaul decentralized autonomous organizations (DAOs). His plan leverages personal artificial intelligence agents to privately cast votes on behalf of users, aiming to scale digital governance and combat low participation and power centralization.
Buterin suggests individuals deploy their own AI models, trained on their values and past communications, to handle the thousands of daily DAO decisions. This addresses the challenge of users lacking the time or expertise for every vote. "We use personal LLMs to solve the attention problem," Buterin stated.
The proposal emphasizes two key areas: privacy of content and anonymity of participants. AI agents will operate within secure environments like multi-party computation (MPC) or trusted execution environments (TEEs) to process sensitive data privately. Zero-knowledge proofs (ZKPs) will ensure voters remain anonymous, proving eligibility without revealing wallet addresses or voting choices. This aims to prevent coercion, bribery, and the influence of large token holders.
These AI stewards will automate routine governance tasks and flag critical issues for human review. To combat low-quality proposals, Buterin suggests prediction markets where AI agents can bet on proposal success, incentivizing valuable contributions and filtering out noise.
Privacy-preserving tools like MPC and TEEs will enable AI agents to assess sensitive data, such as job applications or legal disputes, without exposing it on a public blockchain.
