Aave's decentralized autonomous organization voted near-unanimously on March 23 to advance the protocol's V4 upgrade toward Ethereum mainnet deployment. The Snapshot ballot returned 645,000 votes in favor with fewer than one vote in opposition — a political mandate inside the DAO for the largest ...
"Aave V4 has successfully passed the ARFC stage. Our team has been working hard to bring Aave V4 to mainnet. Next up: final AIP deployment and a safe, controlled, security-first launch." — Stani Kulechov, Founder & CEO, Aave Labs
Aave's decentralized autonomous organization voted near-unanimously on March 23 to advance the protocol's V4 upgrade toward Ethereum mainnet deployment. The Snapshot ballot returned 645,000 votes in favor with fewer than one vote in opposition — a political mandate inside the DAO for the largest architectural overhaul since the protocol's inception. The vote, however, is non-binding; a formal on-chain Aave Improvement Proposal (AIP) with deployed contract addresses and final risk parameters must still pass before activation.
The timing is significant. Aave holds $27.3 billion in TVL — $8.5 billion ahead of second-place Lido — and in February became the first DeFi lending protocol to surpass $1 trillion in cumulative loan originations. Yet the protocol's governance apparatus is fracturing. The Aave Chan Initiative (ACI), its most prominent delegate led by Marc Zeller, announced in March that it will exit the DAO by July, citing concerns that concentrated voting power allows a single entity to pass its own budget proposals over community opposition. V4's deployment now proceeds against a backdrop of technical ambition and institutional stress.
The Snapshot vote ran March 19–23, 2026, framed as an Aave Request for Comment (ARFC) — the first stage in Aave's multi-step governance process. More than 645,000 AAVE tokens voted in favor of advancing V4 to Ethereum mainnet. Fewer than one token voted against. No abstentions were recorded.
This near-unanimity stands in contrast to the contentious "Aave Will Win" funding proposal one month earlier, which cleared its temperature check on March 1 with a slim 52.58% approval (622,300 for, 497,100 against, 64,200 abstaining). The divergence suggests broad alignment on V4's technical direction even as the community remains divided on governance structure and fund allocation.
The Snapshot result does not authorize deployment. The next required step is an on-chain AIP containing deployed contract addresses, final launch parameters reviewed by risk service providers, and binding token-weighted approval. Aave Labs has not published a target date for the AIP, though the cadence of governance activity suggests it could materialize within weeks.
Aave V4 replaces the monolithic pool model used since V1 with a modular hub-and-spoke architecture. The design separates shared liquidity (held in Hubs) from individual borrowing markets (defined as Spokes), each with its own supply and borrow caps, risk parameters, and pause conditions.
The candidate setup under discussion includes three Hubs:
Each Spoke defines a discrete borrowing market. Per-Spoke limits and throttling mechanisms protect protocol solvency by preventing collateral and liquidation behavior in one market from cascading across the system. This is Aave's answer to a problem that has plagued monolithic DeFi lending pools: a single bad collateral type can endanger the entire liquidity base.
According to a risk-centric analysis published on the Aave governance forum, the architecture allows the DAO to onboard new asset types — including tokenized gold, structured yield products, and real-world assets — without exposing existing depositors to the risk profile of those instruments.
Aave Labs reports that V4's codebase has undergone 345 cumulative days of security review, funded by a $1.5 million DAO-ratified security budget. The program combined multiple layers:
The Sherlock contest reported no critical or high-severity vulnerabilities. All identified issues were addressed and re-validated before final reports were published on GitHub. By DeFi standards, this represents one of the more thorough pre-launch security programs executed to date.
The context matters. The Balancer exploit of November 2025, which drained approximately $110–128 million and led to the protocol's corporate dissolution in March 2026, underscored the consequences of security failures. Aave's $1.5 million investment in pre-deployment auditing is a direct response to the market's pricing of security risk.
The ARFC proposal specifies that V4 will go live with a deliberately narrow activation surface: minimal assets, conservative supply and borrow caps, and defined risk parameters set by service providers. The initial asset configuration is expected to include bitcoin, ethereum, and stablecoins, with more niche instruments such as tokenized gold and structured yield products potentially onboarded in subsequent phases.
The DAO retains the ability to lift caps, extend or resize credit lines, onboard additional assets, and configure new Hubs or Spokes once live conditions support it. The phased approach is intended to allow monitoring of liquidity routing, utilization concentration, and credit-line draw behavior before wider exposure.
This is a departure from DeFi's historical pattern of launching with maximum configurability and scaling back after incidents. V4's conservative posture reflects institutional expectations — organizations deploying capital against smart contracts want predictable risk boundaries, not "move fast and break things" deployment philosophy.
The V4 vote's unanimity masks deeper structural tensions. The Aave Chan Initiative (ACI), founded and led by Marc Zeller, announced in March 2026 that it will not renew its engagement with the DAO and will exit by July. ACI has been one of Aave's most active governance delegates and service providers.
The rupture traces to December 2025, when a community member questioned why fees from a CoW Swap integration were being routed to Aave Labs rather than the DAO treasury — a decision made without a governance vote or formal announcement. The dispute escalated through dueling governance forum posts, culminating in the "Aave Will Win" funding proposal.
Zeller's core allegation: approximately 233,000 tokens from three clusters linked to Aave Labs — including a 111,000-token delegation from co-founder Kulechov — swayed the "Aave Will Win" vote's outcome. In Zeller's assessment, "a single entity holds enough voting power to pass its own budget proposals over community opposition," and that "same voting power could cancel any active stream at any time."
ACI's departure follows BGD Labs' earlier exit, marking a second major contributor departure. ACI is seeking to cancel its GHO stablecoin funding stream (ID 100070) ahead of schedule, requesting 120 days' worth of the stream be transferred to ACI's treasury with the remaining balance returning to the Aave Collector.
The governance fracture raises a question the V4 vote cannot answer: whether a protocol with $27.3 billion in TVL can maintain decentralized governance when its founding entity controls sufficient voting power to determine outcomes unilaterally. The technical architecture may be decentralized; the political architecture, according to departing delegates, is not.
Aave's dominance in DeFi lending is quantitatively clear:
| Metric | Aave | Morpho | Compound | |--------|------|--------|----------| | TVL | $27.3B | ~$10B+ | $2.08B | | Market Share (DeFi Lending) | 62.8% | Growing | Declining | | Monthly Fees | $83.3M | ~$21M (est.) | Lower | | Cumulative Loans | $1T+ | N/A | N/A | | Architecture | Hub-Spoke (V4) | Modular/P2P | Monolithic |
Aave surpassed $1 trillion in cumulative loan originations in February 2026, a milestone confirmed by Kulechov. This figure represents aggregate borrowing volume since inception, not outstanding balances — but it signals the protocol's throughput scale.
Monthly fee generation of $83.3 million is approximately four times that of Morpho, Aave's nearest competitor. Against $75.8 million in annualized protocol revenue (after incentive distributions), the protocol captures a meaningful fraction of gross fees.
DeFi lending as a sector has crossed institutional thresholds, capturing roughly two-thirds of the $73.6 billion crypto-collateralized lending market, according to industry estimates. Aave's position at the center of this market makes V4's deployment a systemic event for DeFi infrastructure.
Aave Labs launched Horizon in August 2025 as a permissioned lending market on Ethereum where qualified institutional users borrow stablecoins against real-world asset (RWA) collateral. The platform operates as non-custodial infrastructure with smart-contract-executed borrowing and lending.
Current metrics:
Horizon represents Aave's strategy to extend the protocol's lending infrastructure into traditional finance — a bridge between DeFi's composable liquidity and the regulated capital markets that control the majority of global assets. If V4's hub-and-spoke architecture enables cleaner risk isolation, Horizon's institutional borrowers benefit directly: their RWA collateral sits in a Spoke whose risk parameters are independent of DeFi-native volatile assets in other Spokes.
Aave V4 is the most consequential DeFi protocol upgrade currently in progress. Its hub-and-spoke architecture addresses real structural limitations — risk contagion, liquidity fragmentation, and the inability to safely onboard heterogeneous collateral types — that have constrained lending protocols since the sector's inception.
The security program is substantive by any standard. The governance mandate is clear. The market position is dominant.
But the protocol's political infrastructure has not kept pace with its technical ambition. Two major service providers have departed or announced departures within months. The core allegation — that Aave Labs controls sufficient voting power to determine governance outcomes — remains unresolved. For a protocol managing $27.3 billion in user deposits, the gap between technical decentralization and governance centralization is not an abstract concern.
V4's mainnet activation, likely within the coming weeks pending a final AIP vote, will test whether Aave's architecture can scale to institutional requirements. The governance question — who actually controls a $27 billion protocol — may prove harder to resolve than any smart contract audit.