DeFi United Rescue Plan Secures $100 Million in Pledged Support for rsETH Attack

DeFi United, led by Aave and centered on cross-protocol coordinated rescue efforts, has been established, with multiple protocols joining to participate in the rescue.

Article by ChandlerZ, Foresight News

On April 18, the Kelp DAO rsETH LayerZero bridge was compromised, resulting in the unauthorized minting of 116,500 rsETH (approximately $292 million). The attacker subsequently borrowed assets equivalent to about $190 million on Aave V3. The Arbitrum Security Council froze 30,766 ETH at the chain level related to the incident. In response, DeFi United—a mechanism led by Aave and centered on cross-protocol coordinated rescue efforts—was immediately established with the goal of covering the total shortfall of 112,200 rsETH (approximately $258 million).

As of April 24, multiple protocols have publicly expressed their participation in DeFi United, with combined pledged commitments totaling approximately $100 million; however, most are still in the stages of DAO voting, forum discussions, or earlier preliminary discussions.

The Golem Foundation and Golem Factory have collectively withdrawn 1,000 ETH (approximately $2.3 million) from the treasury to restore rsETH collateral backing and facilitate an orderly resolution for affected users. Golem has stated that it has been closely collaborating with the Aave team, and this contribution is part of DeFi United’s cross-protocol effort to address risk.

Aave founder Stani Kulechov has personally pledged 5,000 ETH (approximately $11.5 million) to support DeFi United. In a tweet, Stani stated that Aave is his life’s work, and the team is fully committed to finding the best possible solutions for users, while working with partners to secure additional funding commitments. This personal contribution does not go through the DAO voting process and is a direct pledge.

The EtherFi Foundation authorizes the deployment of up to 5,000 ETH (approximately $11.5 million) from the DAO treasury as EtherFi’s share in a cross-protocol joint rescue effort to absorb user losses and prevent bad debt in lending markets such as Aave. The program will automatically terminate upon completion of the rescue, reaching the 5,000 ETH cap, or following a subsequent governance vote to end it; any unused funds will be returned to the DAO treasury. If compensation is later obtained through legal proceedings, insurance payouts, or on-chain recovery, recovered funds will be returned to the DAO up to the amount of its original contribution. The Foundation commits to publishing details of the rescue framework and allocation mechanism within seven days of proposal approval.

Lido DAO proposes to contribute no more than 2,500 stETH (approximately $5.75 million) to a dedicated rescue vehicle, with funds bridged via a multisig wallet operated by the Lido Labs Foundation. This allocation will only be released if the rescue fund is fully subscribed and sufficient to fully cover the rsETH shortfall. Lido argues that if only a partial shortfall is covered, depositors in the EarnETH vault would still face exposure to losses of up to approximately 9,000 ETH; therefore, Lido refuses to unilaterally fund a partial rescue structure. The rescue vehicle’s mandate is strictly limited to covering the rsETH shortfall itself and does not extend to secondary effects such as supporting collateral health factors or resetting secondary loss capital. This allocation is separate from the EGG 2026 budget; any unused or reclaimed funds will be returned to the Lido Aragon Agent. The proposal has not yet entered the Aragon on-chain voting phase.

The Mantle proposal suggests that the Mantle Treasury extend a loan of up to 30,000 ETH (approximately $69 million) to the Aave DAO, accruing interest at the LIDO yield plus 1% APR, with a maximum term of 36 months and no penalties for early repayment. The terms require the Aave DAO to provide collateral consisting of at least $11 million in AAVE tokens plus 5% of future protocol revenues. The funds will be deposited into a multisig wallet held by Mantle with senior priority, and voting rights for 130,000 AAVE tokens will be delegated to Mantle. Mantle positions this proposal in the draft as a fixed-income instrument priced at a crisis premium, aiming to incentivize the native deployment of Aave on the Mantle network.

Several other protocols have publicly announced their participation but have not yet disclosed amounts, including Ethena, LayerZero, and INK Foundation.

Based on current commitment levels and the total shortfall, DeFi United has an estimated remaining gap of approximately $50 million. Even if external commitments remain at current levels, the Aave protocol itself still has sufficient buffer. The Aave treasury holds approximately $180 million in assets, and the Umbrella safety module holds approximately $56 million in insurance reserves, totaling approximately $236 million—already covering the upper end of LlamaRisk’s estimated bad debt range of $123.7 million to $230.1 million.

Aave has partially unfrozen the previously emergency-frozen WETH market. The initial liquidity run of approximately $3 billion during the attack has largely stabilized, and the AAVE token has begun to recover after an earlier decline of over 25%. From the perspectives of protocol self-rescue and market sentiment, the systemic risk of the event has moved beyond its most critical phase.

However, Spark’s strategy lead, monetsupply.eth, criticized Aave’s decision to unfreeze the Ethereum core WETH market, calling it ill-considered and primarily beneficial to cycle lenders. He noted that since Aave Ethereum core borrowing rates are capped at 5.15%, while LSTs/LRTs offer higher returns due to discounts and staking yields, arbitrageurs can achieve approximately 45% annualized returns through cycle lending with up to 14x leverage. This strategy will keep the WETH market utilization persistently at 100%, preventing ordinary aWETH holders from withdrawing collateral or refinancing high-cost debt positions into other markets. He further questioned whether the decision was driven more by public relations than sound risk management, effectively worsening the situation for ordinary users already trapped in the market over the past few days.

The actual scale of bad debt that Aave needs to absorb ultimately depends on KelpDAO’s loss-sharing scheme. 0xngmi, founder of DefiLlama, conducted a comprehensive numerical analysis of three possible pathways, each of which would have vastly different impacts on Aave.

Path One: All rsETH holders face proportional loss deductions. KelpDAO applies a uniform 18.5% deduction to all rsETH holders, corresponding to the share of the 116,500 unauthorized rsETH minted within the total circulating supply. Currently, approximately 666,000 rsETH are used as collateral across Aave’s entire network, with positions on both mainnet and L2 largely near maximum circular leverage (95% liquidation LTV). Once the socialized loss is implemented, the 18.5% deduction far exceeds the approximately 5% equity buffer in circular positions, resulting in the complete wipeout of equity for all positions on mainnet and L2, triggering mass liquidations. According to 0xngmi’s estimates, the final bad debt accumulated on Aave amounts to approximately $216 million, or 13.5% of the total collateral value.

The coverage structure for this bad debt consists of approximately $55 million covered by the Umbrella insurance module and an additional $85 million subsidized by the Aave treasury, leaving a remaining shortfall of approximately $76 million. KelpDAO can bridge this gap by borrowing or selling the AAVE tokens held by the Aave treasury, which currently have a market value of about $51 million. This approach distributes the losses collectively among all users, limiting the impact on a single Aave market, but directly reduces the权益 of rsETH holders.

Path Two: Abandon L2 Holders, Protect Only Mainnet. KelpDAO chose to repay only mainnet rsETH holders at par value, treating rsETH on L2 as worthless assets. The value of rsETH collateral on L2 from Aave is approximately $359 million (valued at parity using the current oracle price); if all positions were at maximum circular leverage, bad debt would directly swell to $341 million.

Under this scenario, the Umbrella insurance module cannot cover any L2 bad debt, leaving Aave to rely solely on treasury funds or external borrowing to rescue parts of the market. The rsETH markets on Arbitrum, Mantle, and Base—each suffering the largest losses—are most likely to be abandoned and collapse. While this approach has minimal direct impact on Aave mainnet, it would severely damage the reputation of the L2 ecosystem and could trigger a broader loss of trust across the entire L2 restaking sector. For Mantle, this scenario would result in the direct collapse of its own L2 rsETH market, undermining the credibility of its 30,000 ETH loan proposal.

Path Three: Repay only original holders based on the pre-attack snapshot. KelpDAO attempted to fully reimburse only those who held rsETH prior to the attack, leaving subsequent buyers or transferees to bear their own losses. In theory, this would minimize losses to the narrowest possible scope, but actual implementation is extremely complex: after the attack, funds had already flowed extensively across DeFi protocols, and the nature of lending and liquidity pools involves commingled capital, making it impossible to truly distinguish deposits from different batches.

In terms of specific bad debt calculations, the hacker borrowed $124 million on Aave Mainnet and $18 million on Arbitrum, totaling $142 million; after deducting Umbrella’s coverage, the net loss remains approximately $91 million. This scenario entails the theoretical minimum loss but is nearly impossible to implement technically, and would likely provoke prolonged disputes at both legal and community levels. 0xngmi assesses its likelihood of implementation as extremely low.

Across the three pathways, Aave's net losses range from $76 million, $91 million, to $341 million—a difference of orders of magnitude. KelpDAO’s choice of solution will determine whether Aave must fully utilize its treasury resources.

Rescue funds have been assembled around Aave, but the critical decision on whether the plan can proceed rests with KelpDAO. On the afternoon of April 23, KelpDAO issued a statement reaffirming its core principle that users are always prioritized, noting that over the past four days, the team and partners have made substantial progress along multiple pathways, and publicly thanked the Arbitrum Security Council for the on-chain freeze and SEAL 911 for their early investigative involvement. KelpDAO did not disclose specific details of the plan, timeline, or loss allocation, but committed to continuing to share updates through official channels.

The current DeFi United rescue plan centers around Aave, with the baseline scenario implied by the intended commitment structures of various protocols being Path One. Should KelpDAO pursue an alternative path, the narrative surrounding Mantle’s 30,000 ETH loan, the allocation ratios between EtherFi and Lido, and the entire rescue framework would need to be recalculated.