Ethereum co-founder Vitalik Buterin said on an X post that artificial intelligence could help create efficient models of decentralized governance for informed user experience.
AI agents to fix DAO governance
Buterin posted on X on Sunday, stating that one of the main issues with democratic and decentralized modes of governance, like DAOs, is the “limits to human attention,” as most decisions require domains of expertise and time that the system lacks.
“The usual solution, delegation, is disempowering. It leads to a small group of delegates controlling decision-making while their supporters, after they hit the delegate button, have no influence at all,” he said.
Decentralized Autonomous Organizations or DAOs, let token holders vote on proposals, including funding decisions or protocol upgrades. He proposed a technical solution for the DAOs.
His proposed solution to the concern is to use personal large language models (LLMs) to solve the individual problems. Listing down further, he even suggested a few ways to go about the same, including personal governance agents, public conversation agents, decentralized governance with private information, and others.
Instead of humans manually voting on every DAO proposal, AI agents could act as digital representatives, making governance more scalable, private, and intelligent.
Privacy concerns addressed as Vitalik
At the core of this move, Buterin places privacy as the key. Keeping information private and protecting the users.
In DAO, proposals and discussion could include sensitive information. For content privacy, AI agents would run in secure systems, including multi-party computation (MPC) or trusted execution environments (TEEs), to process private information without exposing it on the public blockchain.
For voter anonymity, Buterin suggested using zero-knowledge proofs (ZKPs), allowing users to prove their eligibility to vote without revealing their wallet address or how they voted.
AI would handle the boring, repetitive parts of governance and alert the users when something major and unusual happens, analyzing all the data and managing votes.
“If the agent is unsure how you would vote on an issue and is convinced it is important, then it should ask you directly and give you all relevant context,” Buterin added.
