DAOs Govern $26B Through Slow Pipelines and Concentrated Delegates
On-chain treasuries exceed $26B, but governance cycles take 14-30 days, quorums are high, and top 10 delegates can pass proposals in 17 of 23 major DAOs.
In the year of our algorithm, DAOs now govern over $26B in on-chain treasuries. Uniswap ($4.8B), Sky/MakerDAO ($3.9B), Optimism ($2.1B), Arbitrum ($1.7B), and Lido ($1.4B) lead the pack [^claim_755]. But the machinery moving this capital is slow, concentrated, and increasingly reliant on off-chain infrastructure—a system that would make a 19th-century railway baron nod in recognition of its bottlenecks and choke points.
The standard governance pipeline runs 14 to 30 days: forum discussion (5-14 days), a Snapshot temperature check, an on-chain vote (3-7 days), then a 2-7 day timelock [^claim_756]. Quorums are high—Uniswap demands 40M UNI for on-chain proposals; Aave requires 80K AAVE for Snapshot [^claim_757]. That protects against hostile takeovers but also ossifies decision-making. Most treasuries hold 60-90% of their value in native tokens, so active management focuses on the 20-40% stablecoin/ETH slice [^claim_758]. Diversifying out of the native token requires governance approval and creates market impact—a liquidity trap dressed in democratic clothes.
The yield stack has shifted toward tokenized U.S. Treasuries. Sky alone holds $2.1B in RWA positions across vehicles like BlockTower Andromeda and BlackRock BUIDL; ENS and Optimism have also deployed into BUIDL [^claim_759]. This embeds new centralization vectors: issuers, custodians, and oracles become critical dependencies. The interface is cold, like a Swiss vault door, but the keys are held by a few hands.
Governance itself is increasingly delegated. Sky uses a service-provider model—Steakhouse, Phoenix Labs, Block Analitica—operating within policy envelopes. Aave, Uniswap, and ENS have treasury committees with Safe multisig authority [^claim_760]. Yet voting power remains concentrated. Uniswap on-chain proposals see only 5-15% participation. An OpenZeppelin review found that in 17 of 23 major DAOs, the top 10 delegates could unilaterally pass proposals [^claim_761]. That creates clear targets for capture, bribe markets, and governance MEV. The yield of compliance is low, but the short-selling of truth is high.
Recent high-stakes votes illustrate the pattern. Aave’s “Aave Will Win” proposal passed with 75% support, redirecting 100% of revenue from all Aave-branded products to the DAO treasury after a dispute over unvoted fee redirection [^claim_762]. Arbitrum’s STEP 2.0 authorized transferring 35M ARB to a multisig for conversion into stable, yield-earning RWAs, with a structured RFP and tokenholder vote on allocation [^claim_763].
Spend authorization is stratified. Routine expenses are batched into grants—Optimism RetroPGF runs about $50M per round, Arbitrum STIP/LTIPP between $40M and $200M, Uniswap grants roughly $40M per year, ENS Public Goods around $1.5M to $3M per quarter. Large strategic deals, like Maker’s $1B+ BlockTower deployment, always go to full votes [^claim_765]. Curve’s veCRV system adds another layer: 7-day windows, 30% quorum with 51% support for some proposals, 15% quorum with 60% for others [^claim_764].
The bottom line: DAO governance is hardening against attacks but creating latency and concentration risk. Reliance on Safe multisigs, Snapshot polling, and outsourced analytics means much of ‘on-chain governance’ is mediated by off-chain infrastructure—creating new points for censorship, data manipulation, and coordination failures. Protocol designers and infrastructure builders should target credibly neutral, verifiable, and cross-chain-aware primitives to reduce these risks. The market is bleeding red like a bruised arm, and the only hedge is to understand the system’s architecture.
Provenance ledger
11 claims web-citedEvery claim below cites a source URL, and each URL was checked for validity before publish. The excerpt shown is the researcher's own summary of the page — it is not re-derived from the source, so it is not a verified verbatim quote. Follow the link to confirm any claim against the original. Citation markers in the text jump here.
[1] As of Q1 2026, DAOs collectively control more than $26B in onchain treasuries, with Uniswap at about $4.8B, Sky/MakerDAO at $3.9B, Optimism at $2.1B, Arbitrum at $1.7B, and Lido at $1.4B in treasury assets tracked by DeepDAO. web-cited
“As of Q1 2026, DAOs collectively control more than $26B in onchain treasuries, with Uniswap ($4.8B), Sky/MakerDAO ($3.9B), Optimism ($2.1B), Arbitrum ($1.7B), and Lido ($1.4B) the largest individual treasuries per DeepDAO's tracker.”
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[2] DAO treasury decisions typically follow a four-stage governance pipeline—forum discussion (5–14 days), off‑chain Snapshot temperature check, on‑chain governance vote (3–7 days), and a 2–7 day timelock—so that a full cycle from proposal to executed treasury transaction usually takes 14–30 days. web-cited
“A community member posts a proposal idea on the DAO's discussion forum… Discussion runs for 5-14 days… An off-chain Snapshot vote tests community support… If the Snapshot passes, the proposal moves to an onchain vote… Onchain voting periods range from 3-7 days… Passed proposals enter a time-lock (typically 2-7 days)… The full cycle from proposal posting to fund release is typically 14-30 days.”
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[3] Quorum requirements for treasury and protocol-upgrade votes in large DAOs are high: Uniswap requires 25M UNI quorum for Snapshot temperature checks, 40M UNI quorum for on‑chain governance proposals to pass, and on‑chain voting windows of 3–7 days, while Aave requires 80K AAVE quorum on Snapshot for temperature checks. web-cited
“Snapshot votes are signed by token holders but don't transfer or commit funds. Quorum and approval thresholds vary; Uniswap requires 25M UNI for quorum on Snapshot, Aave requires 80K AAVE… Onchain voting periods range from 3-7 days. Quorum and approval thresholds for treasury proposals are typically higher than for parameter changes, Uniswap requires 40M UNI for governance proposals to pass.”
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[4] Most large DAO treasuries hold 60–90% of their total value in their own native governance token, so active treasury management and diversification decisions tend to focus on the remaining 20–40% slice held in stablecoins and ETH rather than selling the native token, which would require a governance vote and can materially impact market price. web-cited
“Most DAO treasuries hold the native token but rarely sell it, focusing operational decisions on the stablecoin and ETH allocations… Most DAOs hold 60-90% of treasury value in their native governance token because the initial token allocation dominates. Active treasury management focuses on the stablecoin and ETH slice (typically 20-40% of treasury). Diversifying out of the native token requires governance approval and creates material market impact.”
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[5] The DAO treasury yield stack has shifted heavily toward tokenized U.S. Treasury bills, with Sky (formerly MakerDAO) alone holding around $2.1B of tokenized RWA positions across vehicles like BlockTower Andromeda, Monetalis Clydesdale, and BlackRock BUIDL, while ENS and Optimism have also deployed treasury into BlackRock’s BUIDL fund. web-cited
“The asset mix in DAO treasury yield deployments has shifted markedly toward tokenized US Treasury bills. Sky alone holds $2.1B in tokenized RWA positions; ENS deployed into BUIDL in 2025; Optimism's foundation holds substantial BUIDL position. The shift reflects both yield (T-bills near 4.4% versus DeFi money markets in the 4-6% range) and risk (tokenized T-bills carry less smart-contract risk than DeFi money markets).”
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[6] Governance activity around protocol upgrades and treasury policy is increasingly delegated to professional service providers and committees: Sky uses a service‑provider model (Steakhouse, Phoenix Labs, Block Analitica) operating within governance‑approved policy envelopes, while Aave, Uniswap, and ENS have treasury committees or working groups with Safe multisig authority over parts of the treasury. web-cited
“Sky pioneered this [service-provider model]: governance approves a service provider (Steakhouse Financial, Phoenix Labs, Block Analitica, others), defines a policy envelope… and delegates within-envelope decisions to the provider… Aave, Uniswap, and ENS have established treasury committees or working groups with multi-sig signing authority over a portion of treasury, operating within governance-approved policy.”
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[7] Voter participation in on‑chain governance for major DAOs is low and highly concentrated: Uniswap on‑chain proposals typically see only 5–15% of UNI supply participating, and an OpenZeppelin 2024 review found that in 17 of 23 major DAOs, the top 10 delegates alone held enough voting power to unilaterally pass proposals. web-cited
“Most DAO governance votes have low participation, Uniswap proposals typically see 5-15% of UNI participating. Effective governance often depends on a small set of large delegates. The OpenZeppelin governance review of 2024 found that for 17 of 23 major DAOs, the top 10 delegates controlled enough voting power to pass proposals unilaterally.”
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[8] Aave’s "Aave Will Win" (AWW) proposal passed in April 2026 with 75% support and redirected 100% of revenue from all Aave‑branded products, including Aave App, Aave Pro, and Horizon, into the Aave DAO treasury to resolve a dispute over swap fees being diverted away from governance without a vote in late 2025. web-cited
“In April 2026, the Aave DAO passed the "Aave Will Win" (AWW) proposal with 75% support — described by founder Stani Kulechov as "the most important proposal in Aave's history." The vote redirected 100% of revenue from all Aave-branded products (including Aave App, Aave Pro, and Horizon) to the DAO treasury, resolving a months-long dispute that began in December 2025 when swap fees were redirected away from the DAO without a governance vote.”
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[9] Arbitrum’s Stable Treasury Endowment Program (STEP) 2.0 governance proposal authorized transferring 35,000,000 ARB from the Arbitrum DAO treasury to a STEP multisig to be converted into stable, liquid, yield‑earning real‑world assets, with a structured RFP and timeline where ARB holders vote on allocation after finalists are announced. web-cited
“This vote authorizes transfer of another 35 million ARB to the STEP multisig for the next edition of the Stable Treasury Endowment Program (STEP), to diversify Arbitrum DAOs treasury into real world assets that are stable in value, liquid in conversion and have yield uncorrelated to crypto markets… This program is intended to convert 35,000,000 ARB into stable, liquid and yield earning assets… Submissions open : Feb 20th… Finalists announced : April 20th… ARB holders vote on proposed allocation
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[10] Curve DAO’s on‑chain governance for protocol changes uses veCRV‑weighted voting with fixed 7‑day proposal windows, requires 30% quorum and 51% minimum support for some classes of proposals and 15% quorum with 60% support for others, and enforces that only addresses holding veCRV at the start of the vote can participate. web-cited
“All proposals are open for voting for 7 days and require specific quorum and support thresholds to pass… Requires 30% quorum and 51% minimum support to pass… Requires 15% quorum and 60% minimum support to pass… Voting on proposals always happens on Ethereum. Anyone who already held veCRV when the vote began can participate in voting… You need 2,500 veCRV to create an on-chain governance proposal.”
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[11] Across major DAO treasuries, spend authorizations are stratified: routine operational expenses are usually batched into 6–12 month service‑provider or working‑group grants, grant programs like Optimism’s RetroPGF (~$50M per round), Arbitrum’s STIP/LTIPP (~$40M–$200M per round), Uniswap’s grants (~$40M per year), and ENS Public Goods (~$1.5M–$3M per quarter) are delegated below thresholds, while large strategic deals (e.g., Maker’s $1B+ BlockTower Andromeda deployment or Uniswap Foundation’s $4M Llama acquisition) always go to full governance votes. web-cited
“Core team compensation… Typically authorized in batches via ‘service provider grants’, a single proposal authorizes 6-12 months of recurring payments… Optimism's RetroPGF rounds (~$50M per round), Arbitrum's STIP and LTIPP (~$40M-$200M per round), Uniswap Foundation's grant program (~$40M per year), ENS Public Goods Working Group ($1.5M-$3M per quarter)… Strategic / One-Off Spending… Examples: Uniswap Foundation's $4M acquisition of Llama (2023), MakerDAO's deployment of $1B+ into BlockTower An
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.