Aave governance dispute intensifies as ACI founder publishes 'audit' of Aave Labs ahead of $51M funding vote

Aave (AAVE) governance tensions have escalated again after Aave Chan Initiative and its founder, Marc Zeller, published a lengthy "audit" of Aave Labs’ track record ahead of a vote on what ACI has described as the largest funding proposal in the protocol’s history.

Aave Labs had already put out its own "contributions report" earlier in the day, pitching it as a decade-spanning overview of the team’s work since 2017 and a reference point for tokenholders weighing the upcoming "Aave Will Win" proposal. The post walks through Aave’s evolution from the EthLend ICO to the current multi-chain protocol, and argues that key building blocks — including the pooled-liquidity model, Flash Loans, the Safety Module, and V3’s Efficiency Mode — were conceived and shipped by Aave Labs before the DAO’s modern service-provider stack existed

Zeller, however, took issue with the framing in the comments, arguing the report reads like a “marketing document” and sidesteps the accountability questions he said tokenholders should be asking ahead of a record funding request — namely, what Aave Labs delivered, what it cost, and what the DAO received in return.

"As the Snapshot for the $51M ‘Aave Will Win’ ask drops tomorrow, take a look at our Audit of Aave Labs’ performance and their ~$86M in funding they’ve received to date," Aave Chan said in a post on X, linking to the report on Aave’s governance forum.

In the governance post, Zeller wrote that ACI first published its own transparency report and then applied the same framework — "what did you deliver, what did it cost, and what was the return" — to Aave Labs.

The post then argued that Labs has not published an "accountability report" with a cost-per-outcome breakdown, financial disclosure, or wallet transparency, despite receiving sizable funding over time.

ACI raises questions

At the center of Zeller’s critique is the money. He has put Aave Labs’ "total capitalization" at about $86 million, broken out as $16.2 million from the 2017 ICO, $32.5 million from VC rounds, $31.93 million in direct DAO payments, and roughly $5.5 million in what he called "unapproved" swap fees.

The post also pointed to the founding team retaining 23% of the original LEND token supply, which was later converted to AAVE during the migration, while current AAVE holdings remain undisclosed.

Zeller’s audit has leaned heavily on a case study: Horizon, Aave’s RWA-focused market. He contended that while Horizon has posted attention-grabbing total supply figures, the market’s composition and usage look far thinner once incentive farming and idle positions are stripped out.

Based on his onchain read, the post has pegged Horizon’s total supply at roughly $466 million, with about 69% in stablecoins and only 31% in RWA collateral, dominated by a single asset — USCC — accounting for the bulk of that RWA side. He also has claimed concentration risk across user activity, writing that three positions represented 59% of the pool, including a large RLUSD depositor with zero borrows and a DirectMinter position tied to idle GHO.

Once those pieces are removed, Zeller has argued the “actual” RWA lending market is closer to a $135 million single-issuer collateral base backing a smaller set of stablecoin borrows across a limited number of addresses — a functional market, in his view, but not the diversified RWA venue implied by broader TVL headlines.

The audit also scrutinized Horizon’s economics.

Zeller wrote that cumulative DAO revenue in Horizon’s collector sits around $216,000, while incentives and related costs have run into the millions — including about $4.2 million in Merkl incentives since launch and additional costs tied to GHO’s savings rate, producing what he called a rough “$24 spent per $1 earned” profile.

The post further claimed Horizon’s early governance process triggered backlash — including an initial plan for a new token with 15% allocated to the DAO — and that the proposal ultimately passed after heavy support from a single delegation, which it said accounted for 57% of “FOR” voting power in that vote.

V3 concerns

Beyond Horizon, Zeller’s audit tried to re-litigate a broader charge: that Aave Labs has benefited from the protocol’s current revenue engine while, in his telling, contributing little to its post–V3.0 evolution.

In one of the post’s sharper claims, the ACI wrote that Aave Labs delivered V1, V2, and the initial V3.0 codebase — but that subsequent protocol upgrades and operational output came from DAO service providers, which he credited with the bulk of the revenue growth.

The criticism lands as Aave has navigated an already-heated dispute over governance power and incentives — including a separate controversy around swap fees on the aave.com interface. Zeller’s post alleged that partner-fee revenue that previously accrued to the DAO via ParaSwap referrals shifted in mid-2025 when Labs switched integrations, routing 15 to 25 basis points of fees to what it called a Labs-controlled address without a governance vote.

The post has estimated 933 ETH in mainnet distributions plus additional Layer 2 amounts, with a TokenLogic dashboard cited as confirming roughly $5.5 million in total.

Zeller also tied those claims directly to the "Aave Will Win" push, arguing that a promise to route “100% of Aave Labs’ product revenue” back to the DAO reads less like a concession and more like reversing a disputed redirection.

The Aave DAO standoff also previously sharpened after BGD Labs, the core engineering team behind Aave’s V3 upgrades and much of its recent revenue engine, announced it will exit when its contract expires on April 1, warning that governance frictions had made sustained development increasingly difficult.

Governance

The post also widened the lens to governance mechanics. It pointed to a failed conflict-of-interest and disclosure proposal, and said onchain forensic work traced most “NAY” voting power to a cluster linked to founding infrastructure — a claim it used to argue that the same power bloc can swing major outcomes, including Horizon.

It additionally warned that transferring 75,000 AAVE tokens under "Aave Will Win" would not only represent cash value, but also additional voting power to an entity whose broader holdings are not publicly disclosed, in the audit’s view.

The bottom line is that Zeller’s audit has posed a simple question for tokenholders heading into the $51 million Snapshot: whether the DAO should underwrite a new expansion plan — including proposed products and an accelerated path toward Aave V4 — while, he argues, accountability, wallet transparency, and measurable ROI reporting remain unresolved.

The Block has reached out to Aave Labs for comment.

Updated to include Aave Labs' overview.