Who Built Aave?
Aave is DeFi’s largest lending protocol. $26 billion in value locked. $140 million in annual revenue. 60% of DeFi lending. Today, the question of how it got there became contested on its own governance forum, with the founding company and the DAO’s most powerful delegate publishing competing accounts of the same history. Neither is a neutral party. Both have significant governance power and financial interests in the outcome of the vote now before token holders.
What follows is what happened, sourced from the primary documents.
The Proposal
On February 12, Aave Labs published the “Aave Will Win” framework, a proposal under which 100% of product revenue would flow to the DAO, with Labs requesting $42.5 million in stablecoins and 75,000 AAVE tokens in Year One funding. Total value: approximately $51 million. That is 31.5% of the entire treasury and 42% of non-AAVE reserves, allocated to a single service provider in a single bundled vote.
The framework also proposes ratifying V4 as the protocol’s technical future, pausing new V3 features, and planning eventual V3 deprecation. V4 is currently on testnet. V3 generates all of Aave’s revenue.
Community feedback before the Snapshot vote called for wallet disclosure, a Foundation structure before funding is released, and V3 deprecation gated to V4 adoption milestones. The vote has proceeded without enforceable commitments on any of these.
Two Accounts
On February 25, two posts appeared on the Aave governance forum within hours of each other.
Aave Labs published a contributions report. Their account: every protocol version from V1 through V4, Flash Loans, eMode, the Safety Module, GHO, the frontend, and the brand. Over 570,000 lines of code since 2017. Their position on revenue attribution: the architecture enabling those strategies was their original design, and attributing revenue to any single contributor misrepresents how layered protocol development works.
ACI founder Marc Zeller published a financial breakdown. His figures put Aave Labs’ lifetime capitalisation at $86 million, combining the 2017 ICO, venture capital, DAO payments, and swap fees he claims were redirected from the DAO without a governance vote. He traces 23% of the token supply to 52 wallets he connects to the founding infrastructure. He calculates that Horizon, Labs’ institutional RWA product, has cost the DAO roughly $24 for every $1 earned, citing $4.2 million in Merkl incentives against $216,000 in collector revenue. He catalogues six standalone products he argues failed or remain unprofitable, and claims 98% of V3 revenue came from code shipped by BGD Labs and other DAO service providers rather than Labs directly.
Both arguments have merit. Both parties have incentives.
The Departure
Eight days after the Aave Will Win framework was published, BGD Labs announced that it would not renew its engagement upon expiration on April 1.
BGD built V3.1 through V3.7, Liquid eMode, and significant portions of Aave’s governance infrastructure. Their stated reason for leaving: Labs applied pressure to V3 to promote V4, without collaborating with BGD on V4 development, and imposed what BGD described as “artificial constraints” on V3 improvement. The team whose code generates all of Aave’s current revenue concluded that the environment no longer works for them. They have offered a two-month security retainer for critical incidents. After June, they are gone.
Market Context
Since the brand ownership dispute in December 2025, AAVE is down roughly 32%. In the same period, Morpho, which charges zero protocol fees, is up approximately 42%. Whether governance uncertainty is driving the divergence or whether other factors dominate is hard to isolate. What is clear: Aave generates $140 million in annual revenue with a live buyback, and its token has underperformed a direct competitor by over 70 percentage points in two months.
What This Is Actually About
Strip away the competing numbers. The core dispute is straightforward: Aave Labs originated the protocol and argues they deserve ongoing funding commensurate with that contribution. A coalition of DAO delegates argues that Labs is one service provider among several, and should be held to the same accountability standards as any other.
Both positions are coherent. The tension between them is not a dysfunction. It is a sign that something unusual has happened at Aave.
Most DeFi protocols do not have this problem. Most protocols have a founding team that wears a DAO hat, makes the decisions, and receives the funding. In most cases, the governance is theatre. Aave has something different: a real ecosystem of technically capable, financially independent service providers (BGD, ACI, Chaos Labs, TokenLogic) that can genuinely challenge the founding team’s proposals. That is rare. Building that kind of distributed contributor infrastructure is hard, and most protocols never get there.
The conflict exists because “Aave”, however you want to define it, worked.
The question now is whether the DAO recognises what it has before the vote. BGD’s departure is the clearest signal of what is at stake. If the governance environment becomes one in which the best independent contributors leave because Labs’ roadmap crowds out everything else, the distributed model that makes Aave exceptional may start to unravel.
Sources
Aave Will Win Framework: governance.aave.com/t/temp-check-aave-will-win-framework/24055 (Aave Labs, Feb 12, 2026)
Aave Labs Contributions Report: governance.aave.com/t/aave-labs-contributions-report/24155 (Aave Labs, Feb 25, 2026)
“$86 Million, 23% of the Token Supply”: governance.aave.com/t/aave-labs-86-million-23-of-the-token-supply-and-this-is-their-track-record/24159 (Marc Zeller / ACI, Feb 25, 2026)
BGD. Leaving Aave: governance.aave.com/t/bgd-leaving-aave/24122 (BGD Labs, Feb 20, 2026)
[ARFC] $AAVE Token Alignment. Phase 1 — Ownership: governance.aave.com/t/arfc-aave-token-alignment-phase-1-ownership/23616 (Ernesto Boado / BGD Labs, Dec 16, 2025)
 | A guest post by
|