TL;DR

• Vitalik Buterin says most DAOs have been reduced to token-voting treasuries and are failing their original purpose
• He argues DAOs should power real onchain infrastructure like oracles, dispute resolution, and long-term project maintenance
• Token-based governance is inefficient, easy to capture, and mirrors the flaws of traditional politics
• Privacy tools, better communication systems, and careful use of AI are key to building DAOs that actually work
• Ethereum’s long-term decentralization depends on rethinking DAO design, not abandoning it

Ethereum co-founder Vitalik Buterin is calling for a rethink of how decentralized autonomous organizations are built, warning that the current DAO model has drifted far from its original intent. In a Monday post on X, Buterin said the idea behind Ethereum itself was deeply connected to DAOs. They were meant to be systems of code and rules that could manage resources and coordinate people more efficiently and fairly than governments or corporations. That vision, he argues, has been diluted.

Most DAOs today, according to Buterin, amount to little more than “a treasury controlled by token holder voting.” The design works in a narrow sense. Funds can be moved, proposals can be voted on, and decisions technically get made. But Buterin says that is exactly the problem. The model has been copied everywhere despite being “inefficient, vulnerable to capture and fails utterly at the goal of mitigating the weaknesses of human politics.”

Instead of improving on traditional governance, many DAOs have recreated it, complete with whale dominance, low voter participation, and decision-making that favors short-term incentives. That has led to growing cynicism around DAOs as a concept, even among people who once believed strongly in them. But Buterin’s takeaway is not that DAOs should be abandoned. It is the opposite. He argues that Ethereum needs more DAOs, but they need to be designed for specific, difficult problems rather than generic treasury management.

Why Ethereum Needs DAOs for Oracles, Courts, and Shared Infrastructure

Buterin’s core argument is that DAOs should exist to solve infrastructure problems that cannot be handled well by markets or companies alone. One of the first examples he points to is oracles. Oracles are systems that feed external information, like prices or real-world events, into blockchains. They are critical for stablecoins, prediction markets, insurance protocols, and much of DeFi. But Buterin says current oracle designs are deeply flawed.

If an oracle relies on token voting, large holders can influence outcomes on subjective questions. And because a token-based oracle cannot cost more to attack than its market value, it cannot safely secure large amounts of capital without extracting ongoing fees. If the oracle relies on human curation instead, decentralization suffers. As Buterin puts it, “The problem here is not greed. The problem is that we have bad oracle designs.” Bootstrapping better ones, he argues, is not just a technical challenge but a social one, which is exactly where well-designed DAOs should come in. 

A similar issue appears in onchain dispute resolution. Advanced smart contracts, especially in areas like insurance, require systems that can judge subjective outcomes. Was an event legitimate? Did a claim meet the agreed conditions? These are not questions that simple code can answer. But they are also difficult to resolve fairly with today’s governance models. Buterin also highlights the need for DAOs to maintain shared lists. These could include registries of known scams, verified contract addresses, standard interfaces, or trusted applications. Today, much of this information lives in databases or informal community channels. A properly structured DAO could maintain these lists in a transparent, accountable way.

Beyond that, he argues DAOs could help spin up short-term funding efforts quickly, without the overhead of forming legal entities. They could also provide long-term stewardship for projects after founding teams step away, ensuring useful infrastructure does not simply disappear due to lack of maintenance. Taken together, the picture is clear. Buterin is not asking DAOs to vote on everything. He is asking them to do fewer things, but to do those things well.

Privacy, Decision Fatigue, and the Next DAO Stack

For these ideas to work, Buterin says two fundamental problems must be addressed: privacy and decision fatigue. 

Without privacy, governance turns into a social performance. Votes become signals, alliances form in public, and outcomes are shaped by reputation games rather than honest judgment. Buterin has previously warned that transparent governance can sometimes be worse than opaque systems because it amplifies social pressure. Decision fatigue is the second issue. When participants are asked to vote frequently, engagement drops over time. People stop staying informed, not because they do not care, but because the cognitive cost becomes too high. Early enthusiasm fades, and governance becomes dominated by a small, persistent minority.

To deal with this, Buterin points to modern tools that could enable a new generation of DAO design. Zero-knowledge proofs can allow private voting while preserving accountability. Better communication platforms can help communities find consensus without endless proposal churn. He also sees a role for AI, but with strong caution. AI should not replace human judgment or run DAOs autonomously. Instead, it can help summarize discussions, surface trade-offs, and reduce the mental load on participants. In his words, AI should “scale and enhance human intention and judgement, rather than replacing it.”

Buterin stresses that governance is not just about smart contracts. The communication layer matters just as much. Forums, voting tools, and coordination platforms should be treated as core infrastructure, not side features. A simple multisig paired with strong consensus-finding tools, he notes, can outperform far more complex governance systems if designed well.

Closing Thoughts 

The broader message is that DAO design needs to mature. Projects building oracles, governance frameworks, or onchain courts should see governance as half the work, not an afterthought. The same decentralization and robustness that protect Ethereum’s base layer need to extend to the systems built on top of it. For Buterin, DAOs are still a central piece of Ethereum’s future. But only if they evolve beyond token voting and start doing the hard work they were originally meant to do.

READ MORE:  COTI Earn Explained: The Biggest Reward Opportunity You Don’t Want to Miss