Buterin recently noted that <\/span><\/span>DAO voting has a serious participation problem, and his solution sounds straight out of the future. Instead of users manually voting on every proposal, he\u2019s proposed letting personal AI agents do the job.<\/p>\n The idea is that each user would have their own AI model, trained on their messages, opinions, and past decisions. This AI would then vote on DAO proposals exactly how the user would.<\/p>\n \u201cMaking good decisions often cannot come from a linear process of taking people\u2019s views that are based only on their own information, and averaging them (even quadratically).\u201d<\/p>\n<\/blockquote>\n Most people don\u2019t have the time or expertise to vote on thousands of complex decisions. As a result, voting power often ends up concentrated with a few large token holders. Buterin stated,<\/a><\/p>\n \u201cThe usual solution, delegation, is disempowering\u201d<\/p>\n<\/blockquote>\n What\u2019s interesting is that this comes a month after Buterin ripped into DAOs<\/a> for low participation and growing centralization.<\/p>\n \u2026is making sure users don\u2019t have to give away privacy for convenience.<\/p>\n These AI agents would run inside secure systems, like trusted execution environments and multi-party computation. This would allow them to process sensitive information without exposing it publicly.<\/p>\n\n
\n
A key part of the idea\u2026<\/strong><\/h2>\n
![\"buterin\"](\"https:\/\/ambcrypto.com\/wp-content\/uploads\/2026\/02\/Screenshot-2026-02-22-142511.jpg\")