Aave's Winter of Discontent
SEC investigations end. Internal wars begin. And nobody won, at least not yet.
Aave spent four years defending itself from regulators, burning time and resources to prove decentralized finance could work without breaking securities laws.
December 16th, 2025 brought victory: the SEC closed its probe without enforcement action.
That same day, former Aave Labs CTO Ernesto Boado posted a measured proposal to the governance forum asking whether the Aave DAO should own aave.com, the Aave Twitter handle, and the brand IP, assets currently controlled by founder Stani Kulechov's private company, Aave Labs.
The proposal came after weeks of tension over $10 million in annual swap fees that suddenly stopped flowing to the DAO treasury and started flowing to Labs instead.
Standard governance process: Forum discussion, community alignment, then vote.
What happened next would become known as Phase 1.
Five days later, on December 21st, Aave Labs escalated Ernesto's proposal to Snapshot - without his knowledge, without his consent, with his name still attached.
Voting period: December 22-25, ending on Christmas Day.
AAVE dropped 25% as the community realized what was happening.
Polymarket's odds that the token alignment proposal would pass crashed from 65% to 15% in a matter of 24 hours, from December 21st to the 22nd.
The proposal's own author urged everyone to abstain in protest.
The vote would fail on Christmas Day with 55% voting no, 3.5% voted yes, and 41% abstained in protest.
Within a week, Stani would offer revenue sharing. Marc Zeller would demand Phase 2. The token would recover 10% but remain down 20% from where it started.
December 16th freed Aave from regulators. December 21st started a civil war. January 2nd opened negotiations for Round 2.
When freedom from Washington becomes a license to fight over who actually owns what, does decentralized governance provide an answer - or just a more expensive battlefield?
To peel back the layers of complexity, let’s start with how Phase 1 rolled out.
December 4th, 2025, Aave Labs announces a partnership with CoW Swap to provide an improved swap experience across aave.com.
Better execution, improved MEV protection, enhanced user experience - standard upgrade language that usually means nothing to anyone except the engineers implementing it.
Except this time the math told a different story.
ParaSwap had been generating referral fees through a partnership introduced in June 2022. Those fees flowed directly to the Aave DAO treasury - roughly $1.1 million in 2025 alone.
CoWSwap worked differently. The integration came with 15 to 25 basis points in frontend fees. On-chain analysis showed approximately $200,000 per week in ETH distributions flowing to a single address - but not the DAO treasury.
Seven days after the integration went live, EzR3aL - Orbit delegate and the largest independent voice in Aave governance - published his findings on the forum.
He traced the receipts through test transactions on Ethereum and Arbitrum. The appCode clearly indicated the Aave v3 interface widget.
The recipient address: 0xC542C2F197c4939154017c802B0583C596438380
"45.99 ETH have been transferred, with a value of approx. 152k$ at the time of writing this post. It's worth mentioning that this is only the partnerFee received on Ethereum Mainnet."
He ran the same test on Arbitrum. Same address receiving fees. Same pattern on every chain where CoWSwap supported the Aave frontend.
The math was simple: Weekly transfers of roughly $200,000 heading to addresses controlled by Aave Labs, not the DAO treasury.
"A loss to the DAO over 365 days seen by at least over $10m, assuming a transfer of only $200k each week."
Aave Labs responded: “The interface is operated by Aave Labs and sits entirely outside the protocol the DAO stewards. This has always been the separation of responsibilities. In practice, this means the DAO funds protocol development and approves changes to the protocol’s smart contracts on-chain. Aave Labs funds, builds, and maintains its own interface.”
“Any monetization applies only to accessory features such as collateral swaps and does not touch the core economics of the protocol, including borrow interest or other protocol-level flows. That separation keeps the protocol neutral and avoids the kind of centralization that can emerge when a product layer captures value from the core system itself.”
Stani Kulechov went further in forum replies: ParaSwap created surplus when execution beat quotes, and that surplus was donated to the Aave treasury. Roughly 200-250k per quarter. But it could have been refunded to users or gone to Labs "given its not a protocol related feature, it simply swaps on the Aave Labs application."
The donation was voluntary. Once Labs built new CoWSwap adapters and changed routing, that surplus naturally disappeared.
Marc Zeller saw it differently. Every service provider on the Aave DAO payroll has a fiduciary duty to token holders. For years, there had been a tacit understanding: the DAO lends its brand and intellectual property to Labs in exchange for Labs directing interface revenue back to the treasury. That wasn't voluntary charity - it was the deal.
"We've been fooled in considering this a natural alignment, and we acknowledge the new reality."
The debate crystallized around a single question: Does building the frontend give Labs the right to monetize the Aave brand indefinitely, or does the DAO - having funded years of development, paid for the recent rebrand, and owned the underlying protocol - deserve those fees?
Aave Labs pointed out users can interact directly with contracts, self-host their own interface, or the DAO could create a DAO-run interface if desired. Nothing is exclusive to the Labs interface.
Critics countered that charging the fee is only possible because the Aave brand is widely known and accepted - a brand that the Aave DAO has paid for. Users of the protocol tend to use app.aave.com because of brand identity, and the Aave visuals belong to the Aave DAO.
Labs didn't build app.aave.com in a vacuum - it built it on top of infrastructure the DAO paid for, using a brand the DAO made valuable.
Both arguments had internal logic. Both sides claimed alignment with decentralization principles.
But only one side was collecting the checks, so whose version of "decentralization" actually gets enforced?
Competing Visions
Three paths forward. Someone's vision had to lose.
December 16th, 2025. The same day the SEC closed its investigation.
Ernesto Boado, co-founder of BGD Labs and former Aave Labs CTO, posted "[ARFC] $AAVE token alignment. Phase 1 - Ownership" to the governance forum.
The proposal asked a deceptively simple question: Should AAVE token holders have explicit control over the project's brand assets?
Domains including aave.com and all subdomains. Social handles, Aave on Twitter, Discord, Instagram, everything using the Aave name. GitHub organizations, Aave and Aave-dao. NPM namespaces. Naming rights for who can call their products "Aave" something.
All of it currently is stewarded by various entities, primarily Aave Labs and BGD Labs, under informal arrangements that had worked fine until $10 million in annual fees suddenly changed hands.
Ernesto's proposal called for transferring these assets to a DAO-controlled legal vehicle with anti-capture protections and enforceable recourse if brand assets get misused or withheld.
"This is not a discussion suggesting that Aave Labs should not be a contributor to the DAO, or that it lacks legitimacy or capability to do so. Those are totally independent topics."
"I think without the DAO having full and exclusive control of all its assets, there is no future for the idea of a decentralised autonomous organisation where multiple parties contribute based on merit."
Ernesto included disclaimers about his independence, his stake as an AAVE holder, his company BGD Labs' relationship with Aave Labs as service-provider-to-service-provider.
Standard governance procedure: Post proposal, gather feedback, build consensus, then move to vote.
Marc Zeller laid out the case for why that wasn't just philosophy.
The entity formerly known as "Aave Companies" rebranded to Avara around 2021-2022, a neutral corporate name that explicitly reduced association between the private company and the protocol. Focus shifted toward separate ventures, most notably Lens and other product initiatives.
Meanwhile, the bulk of day-to-day execution fell to DAO service providers: BGD Labs, Chaos Labs, LlamaRisk, TokenLogic, and ACI. Risk management, parameter updates, governance coordination, technical stewardship, strategic integrations.
They weren't just maintaining the protocol. They were winning deals that drove real revenue. EtherFi and the LRT wave made Aave a primary venue for liquid staking borrowing.
Plasma deployment became the second-largest network for Aave. MetaMask and Consensys integrations drove user flow. Institutional partnerships with Kraken, institutional-grade distribution paths.
Much of the original talent now contributes through independent, DAO-aligned entities. Ernesto left as CTO to found BGD Labs. Andrey, a critical front-end contributor, went independent. David founded Catapulta to deliver DAO infrastructure.
Zeller himself was a day one employee of Aave (not ETHLend), vesting expressed solely in token terms, no Avara equity. "My incentives have stayed consistent: I have concentrated my efforts on the Aave protocol and AAVE token because that was the vision we signed up for."
Jordan, Aave co-founder, laid out the pattern: "Several founders of the most important service providers originally came from Aave Labs, and left specifically to work for the protocol and the DAO."
The DAO was the engine. Brand assets were the storefront. One side kept the engine running while the other side controlled the signage.
Jordan framed the fundamental tension: "There is eventually a fundamental difference between those who saw DAOs as a regulatory shield in uncertain times and those who saw them as the end state of coordination and transparency."
Twenty-four hours later, someone took the gloves off.
December 17th, tulipking posted "Aave Improvement Proposal: The Poison Pill."
Where Ernesto asked politely, tulipking demanded everything.
The Aave DAO should immediately sue Aave Labs for full ownership of all code, intellectual property, and brand. Claim 100% of Labs' equity. Reclaim all past revenue generated from Aave Labs and Aave-branded products.
"This is not a negotiation; it is a declaration of sovereignty. The DAO owns the protocol, the brand, and the future. Aave Labs exists only at the pleasure of the DAO."
Suddenly Ernesto's measured approach looked like the compromise position between status quo and total war.
On December 16th, Stani published his response: a long-form vision document titled "The Master Plan."
Aave V4, unified liquidity architecture, institutional adoption at scale. Horizon for tokenized real-world assets.
The Aave App as a consumer gateway with zero-fee fiat on-ramps covering 70% of global capital markets. Moving the next trillion dollars in assets onchain. Onboarding the protocol's first million users.
The timing wasn't subtle. Forum erupting over who owns the brand? Here's why it matters that the builder stays in control.
Stani also made this point clear: "I've seen a lot of the discourse within the DAO forum. Let me be very clear, no one cares about Aave more than I do."
Three visions in forty-eight hours. Ernesto's structured transition to DAO control. Tulipking's scorched-earth absorption of Labs. Stani's Master Plan that required keeping the current power structure intact.
Standard procedure would have meant weeks of forum discussion, iterative refinement, building consensus, maybe splitting Boado's proposal into phases with clearer implementation details.
Instead, the clock started ticking toward December 21st, and someone decided talking was over.
When competing visions drop within two days and nobody's backing down, does more discussion actually solve anything, or just give everyone more time to pick sides?
The Apparent Hijacking
When following the rules means breaking trust.
December 21st, Aave Labs submits Ernesto Boado's proposal to Snapshot.
Five days after it was posted. While discussion was still active on the forum. Without notifying the author. Without his consent. With his name still attached as if he approved.
Voting opens December 22nd at 7:40 PM. Closes December 25th at 7:40 PM on freaking Christmas Day.
What kind of Grinch would do this?
Ernesto was completely blindsided:
"To be very clear: This is not, in ethos, my proposal. Aave Labs has (for whatever reason) unilaterally submitted my proposal to vote in a rush, with my name on it, and without notifying me at all. If asked, I would not have approved it."
"It was not my intention to submit the vote while the community was still having a healthy discussion around it, with valuable points appearing continuously. It breaks all codes of trust with the community."
"Public governance is supposed to be for, even if hard sometimes, open discussion. Trying to rush a vote is disgraceful."
Forum posts from December 16th through December 21st show active engagement. Delegates raising questions about implementation mechanics. Discussions about legal structures. Debates over whether the proposal needed more specificity before moving forward. New perspectives appear daily.
Stani's response: "The discussion has been going over the past 5 days already with various opinions and takes, a timeline set on the ARFC temp check. The Snapshot is in compliance of the governance framework."
Stani continues: "Other SPs (for example ACI) brings proposals into voting so there is no new precedent. People are tired of this discussion and getting into a vote is the best way to resolve, this is governance end of the day."
Marc Zeller saw it differently: "We acknowledge Aave unilaterally escalated the proposal to Snapshot without resolving discussion, without clear consensus, and without consent from Ernesto."
"We've posted our position in response to this unprecedented interference in the DAO governance process. Worst outcome that was entirely preventable."
The timing made it worse. December 22-25, spanning through Christmas Eve and ending on Christmas Day.
Major institutions close for holidays. Delegates traveling. Coordination capacity at seasonal lows.
Zeller pointed to another factor: "A recent wave of new delegations with significant voting power" that appeared shortly before the vote. The combination, "adds to the perception of a rushed escalation optimized for outcome rather than legitimacy."
Simo, Protocol Growth at Aave Labs, pushed back on the holiday objection: "The claim that 'discussion was still evolving' is just false. Minimal input. Largely repetitive. Same arguments. Same narrative."
"DAOs stop in December? Pause for Christmas, Easter, Summer? Important decisions don't become less important because it's mid-Dec."
Both sides had points. Five days did generate 100+ replies. The governance framework technically permits escalation at that threshold. Other service providers have moved proposals forward similarly. Governance doesn't actually pause for holidays anywhere in the docs.
But Ernesto didn't consent. His name sat on a proposal he actively opposed advancing. The author calling his own proposal "disgraceful" in its current form creates a legitimacy problem no framework technicality can fix.
Ernesto told supporters: "If you don't agree with the legitimacy of the proposal, Abstain or don't participate. I will be creating my own when the proposal has had proper time to discuss and digest."
Marc Zeller and the ACI formally called for ABSTAIN votes: "Following the proposal author's recommendation, we will cast an ABSTAIN vote and invite the community to do the same."
Polymarket opened a market immediately. Odds collapsed from 50% to 4% within days as the author condemned his own proposal and major delegates called for abstention.
The Market didn't believe in the process. The community didn't believe in the timing. The author Ernesto didn't believe in his own proposal anymore.
But the vote was live, and to some it could appear that someone was counting on confusion and holiday absences to deliver a specific outcome.
When governance by consent becomes governance by technicality and the proposal author becomes the loudest dissenter, what exactly is the vote measuring?
The Market's Verdict
When governance breaks down, people vote with their wallets.
Markets don't wait for governance drama to resolve. They price in chaos immediately.
December 16th, SEC announcement hits: AAVE rallies 3%. Four years of regulatory uncertainty lifting should be bullish. The Token touches $187.85 as traders process the news.
The same day, Ernesto's proposal drops and the price holds.
Later that same day, December 16th, Stani drops his Master Plan for Aave.
December 17th, the poison pill arrives. Price wobbles but stabilizes just under $180.
December 21st, the forced Snapshot escalation happens. Aave’s price starts to tank.
AAVE fell roughly from $178 to $159 in a single 24-hour period. Down around 10% while crypto Twitter processed what just happened, a proposal author disavowing his own work, major delegates calling for abstention, Polymarket odds collapsing.
Trading volume told the story: Trading saw more than a 220% surge in volume.
One whale moved $37.8 million in AAVE at a loss of ~$13.75 million.
Others followed, exits compounding exits.
Full damage from December 16th through December 23rd: Down 18% in seven days.
Peak to trough: 25% drawdown. AAVE was hovering around $150 as of December 23rd, market cap billions lighter, while the vote that triggered the selloff hadn't even started yet.
Polymarket provided real-time sentiment tracking. Odds for proposal passage started around 50-50, maybe a slight edge to YES given community anger over the CoWSwap fees.
December 18th, the odds got as high as 78.5%.
December 21st, the same day as the forced escalation announcement. The odds were having an identity crisis - started the day around 14%, peaked at 64.5%, then hovered around 25% by the end of the day.
December 22nd, Ernesto's condemnation went public the day before, Zeller's abstain campaign was live, the bottom fell out, the odds dropped from 24% to 4%.
By December 23rd, the writing was on the wall - the chances of the proposal passing hovered around 4%.
Market interpretation: This vote isn't legitimate. Either it fails outright, or the abstain campaign succeeds and nothing resolves.
Either way, governance is broken and AAVE holders are watching the protocol tear itself apart.
Tulip King summarized the sentiment: "It seems like the Labs is trying to rush a conclusion to the issue before the DAO can coordinate, imo if this proposal fails AAVE should go to zero."
Translation: If the DAO can't assert control over its own brand and revenue, what value does the governance token actually represent? You're holding a vote in a system where the founder controls everything that matters - brand, frontend, institutional relationships - and you're just along for the ride.
Duo Nine pointed to alternative models: "Maybe they need to look at Hyperliquid, where 99% of revenue goes to HYPE buy-backs. The team holds and is paid in HYPE. Everyone wins. Why can't Aave Labs do the same?"
Hyperliquid's structure aligns team and token holders through shared upside. The team gets paid in HYPE. 97% of trading fees flow to token buybacks. Everyone benefits from protocol success equally.
Aave's current reality creates a classic "principal-agent scenario". Labs treats the frontend as a "separate product" for monetization. The DAO governs on-chain parameters. Brand value benefits the agent. Token holders watch $10 million annually redirect away from the treasury.
SEC timing made the irony complete.
December 16th, morning: "After four years, we are finally ready to share that the SEC has concluded its investigation into the Aave Protocol. DeFi will win."
December 16th, afternoon: Proposal drops questioning whether Aave DAO and AAVE token holders should regain full control over Aave’s brand, naming rights, and associated assets.
December 21st: Founder's company forces vote on that question without author consent.
December 22nd: The market may have made the decision for them.
Victory over external regulators, immediately converted into internal warfare. Freedom from Washington meant freedom to fight over who actually owns what, and markets priced in the obvious answer: nobody knows, and that uncertainty is expensive.
An Aave Labs spokesperson tried framing the discord positively: "Discussions reflect a maturing ecosystem working through questions of structure, representation, and expectations as it scales."
Markets heard: "We're having a very public fight about fundamental ownership questions we should have resolved years ago, and it's happening via forced votes during Christmas."
Price action doesn't lie. Down 25% from peak while TVL stays stable at $35 billion, while fees stay strong at $885 million annually, while the protocol functionally works perfectly.
When regulatory freedom becomes the starting gun for an internal civil war, does it matter who wins if the token bleeds out before the vote even closes?
At the Crossroads
As Phase 1 ends. Phase 2 begins. And both sides claimed victory - which means nobody really won.
Christmas Day, the vote closed.
Final tally: 55% opposed, 41% abstained, 3.5% in favor.
AAVE traded at $155, down 22% from where the controversy started. Neither side conceded.
Aave Labs: Proposal failed, status quo preserved, move on.
Marc Zeller: "Despite an unfair timeline and every practical disadvantage stacked against the DAO, participation broke records. That is not a defeat for decentralization. It is the opposite of apathy, and that is exactly what a healthy DAO should look like."
He immediately signaled what came next: "When a mature, legitimate vote is re-run with the author's consent and a complete discussion cycle, participation should be even higher."
Translation: This was Round 1. Get ready for Round 2.
One week later, Stani blinked first.
January 2nd, 2026, he posted "How AAVE will win" to the governance forum.
The olive branch came wrapped in strategic vision. Aave faces a crossroads. DeFi lending alone won't sustain growth. The protocol must expand into real-world assets, institutional lending, consumer products. $500 trillion in global financial assets waiting to be tokenized.
Aave V4 would enable it all. Hub-and-spoke architecture. RWA integration through Horizon. The Aave App targeting millions of users. Moving the next trillion dollars onchain.
Then came the commitment that mattered:
"Given the recent conversations in the community, at Aave Labs we are committed to sharing revenue generated outside the protocol with token holders. Alignment is important for us and for AAVE holders, and we'll follow up soon with a formal proposal."
On branding and IP: "With respect to branding, we will work toward a structure in our upcoming proposal that supports this long-term vision with sufficient guardrails for the DAO and Aave token holders."
Revenue sharing. Brand guardrails. Both coming "soon" via "formal proposal."
The token jumped 10% to $166. Markets liked hearing the word "alignment" even if details remained vague.
Marc Zeller responded the same day on the forum: "Vague statements without clear and concise commitments should be discarded as performative."
He acknowledged the move as progress - "A clearer willingness to discuss substance is the right direction." But reminded everyone what just happened: "Since this dispute became public, roughly $500M of $AAVE market cap has been erased. Correlation is not causation, but markets price uncertainty, and the lack of direct engagement on core questions amplified that uncertainty."
The core demand hadn't changed:
"Aave's strategic brand assets and gateways should belong entirely to a DAO-owned vehicle. Stewardship can be delegated back to Avara under a clear operational mandate and security standards."
Ownership to the DAO. Stewardship can be licensed to Labs. But ownership stays with token holders.
Then came Marc Zeller’s Phase 2 requirements list:
What gets transferred: Domains, handles, naming rights, trademarks, and related brand assets.
What gets licensed back: Scope and duration for day-to-day use.
Operational standards: Clear metrics for stewardship performance.
Guardrails: No more unilateral monetization (e.g., surprise CoWSwap fees).
Enforcement: If commitments get breached, the DAO has recourse.
Stability: Structure survives even if corporate priorities shift.
Not principles. Specifics. Not promises. Enforceable agreements.
"ACI is ready to help convene critical stakeholders for off-chain working sessions and translate the output into a clear Phase 2 proposal for the community to review."
Stani offered revenue sharing and brand "guardrails."
Zeller demanded ownership transfer with licensed stewardship and enforceable recourse.
The gap between those positions? That's Phase 2.
When one side offers to share profits while keeping control and the other demands ownership transfer with licensed operation, whose version of "alignment" wins - or do they just keep fighting until the next crisis forces another compromise?
AAVE token holders asked for decentralization and got coal in their stockings instead.
The Grinch didn't steal Christmas from Whoville this year - he just moved the governance vote to Christmas Day and called it framework compliance.*
Aave spent four years proving to regulators that DeFi could work, only to spend December 2025 proving to token holders that governance might not.
The protocol functions perfectly - $40 billion in TVL, $885 million in annual fees, 59% market dominance in DeFi lending - but none of that matters when the founder controls the brand, redirects the revenue, and forces votes without author consent.
Markets priced in the answer: The Aave token was down 25% in six days, Polymarket odds were at 4%, with whales exiting at losses.
The vote failed, Stani offered revenue sharing, Marc demanded ownership transfer, and the token recovered 10% but remains down 20% overall.
Phase 2 begins now, with $500 million in market cap already burned as the entry fee for whatever compromise - or civil war - comes next.
When the protocol works perfectly but governance implodes every time real money's at stake, are token holders governing anything - or just renting seats at someone else's table while the founder decides who gets fed?
REKT는 익명 작성자들에 의한 공공 플랫폼이며, REKT에 작성된 관점이나 내용에 대해서 그 어떤 책임도 지지 않습니다.
기부 (ETH / ERC20): 0x3C5c2F4bCeC51a36494682f91Dbc6cA7c63B514C
disclaimer:
REKT는 당사 웹 사이트의 익명의 작성자 또는 REKT에 의해 게시되거나 관련된 서비스에서 게시되는 콘텐츠에 대해 어떠한 책임도 지지 않습니다. 당사는 익명 작성자들의 행동 및 게시물에 대한 규칙을 제공하지만, 익명의 작성자가 웹 사이트 또는 서비스에 게시, 전송 혹은 공유한 내용을 통제하거나 책임지지 않으며, 귀하가 웹 사이트 또는 서비스에서 직면할 수 있는 불쾌함, 부적절함, 음란함, 불법 또는 기타 해로운 콘텐츠에 대해서도 책임을 지지 않습니다. REKT는 당사 웹 사이트 또는 서비스 사용자의 온라인 또는 오프라인 행위에 대한 책임을 지지 않습니다.