The stability of the Aave v3 protocol is at a critical point following the exploit of Kelp DAO on April 18. The USDC pool on Ethereum has operated at 99,87% utilization for four consecutive days. With barely 3 million dollars in available liquidity against almost 1.9 billion in deposits, users are encountering serious difficulties withdrawing their funds.
Faced with this congestion, Circle has submitted an emergency proposal to readjust the protocol's risk parameters. The goal is to tackle the prolonged illiquidity and guarantee user withdrawals, avoiding a domino effect that would trigger instability across the DeFi sector.
Delphi, the first prediction market that pays you to create bets
The rsETH event and the market paralysis
This current crisis is not the product of a direct flaw in USDC's design, but of a structural shift in user behavior after the Kelp DAO exploit.
According to internal reports and governance data, incremental loan flows of approximately 300 million dollars aimed at extracting trapped liquidity have been identified.
Liquidity providers holding deposits in frozen markets or in markets with loan-to-value (LTV) ratios cut to zero are using other collateral, such as WBTC, to borrow USDC and exit the ecosystem through decentralized exchanges (DEX).
This dynamic has left the USDC pool on Aave v3 practically locked at its maximum utilization. In fact, by April 22, 2026, utilization stood at 99.87%, leaving available liquidity of barely 3 million dollars, a negligible figure for a market managing close to 1.89 billion dollars in deposits.
Being at the "ceiling" of the utilization curve, borrowing interest rates have remained flat at 14%, a level that has proven insufficient to discourage borrowers or attract new capital organically.
Technical proposal: Readjusting Slope 2 and optimal utilization
To resolve this imbalance, the [ARFC] proposal suggests aggressive intervention in Aave's Interest Rate Model (IRM).
The central goal is to transform the "post-kink" region of the curve into a functional price-discovery range, allowing the market to rebalance through the cost of capital rather than through withdrawal queues.
The proposed changes are divided into two critical phases:
Increasing the Slope 2 parameter
The Slope 2 parameter defines the speed at which interest rates rise when utilization exceeds the optimal point. The proposal sets out to raise the Slope 2 target from the current 10% to 50%, with an immediate intermediate step to 40% through action by the Risk Stewards.
With this, the maximum interest rate at 100% utilization would rise from the current 14% to 53.5%.
Under these stress conditions and with that interest incentive, capital from institutions and sophisticated users should be attracted within hours, restoring the pool's liquidity.
Reducing Optimal Utilization (U*)
The inflection point, or U*, is the utilization level from which interest rates begin to rise exponentially. The proposal suggests reducing this value from 92% to 85%.
This measure seeks to create a broader liquidity buffer (15 percentage points versus the current 8) to absorb demand shocks without locking up the market.
Kelp points to the network and LayerZero to North Korea in the 292 million theft
Governance measures and risk control
In addition to the numerical adjustments, the proposal advocates more direct, manual management during this period of volatility. It has requested pausing the Slope 2 Risk Oracle for USDC.
This oracle, tasked with automating parameter adjustments, showed delays in its response during previous incidents, such as the WETH utilization spike in February 2026, when it took up to 43 hours to react to a liquidity gap.
Execution is planned in stages to guarantee agility:
- Immediate action: Execution of intermediate parameters (Slope 2 at 40%, U* at 87%) through the 2/2 multisig of the Risk Stewards (LlamaRisk and Aave Labs).
- Governance ratification: On-chain vote within 5 to 7 days to set the final targets of 50% and 85% respectively.
- Emergency safeguard: Should the risk managers fail to act, the protocol's 5-of-9 Emergency Guardian multisig remains in place as a contingency.
A proposal unlike any other
It is essential to note that this proposal for USDC is diametrically opposed to the action recently taken with the WETH pool. While for WETH the interest rate curve was "flattened" to avoid mass liquidations of positions backed by rsETH, for USDC the aim is to "steepen" the curve.
Kalshi and Polymarket storm the perpetual futures market
This apparent contradiction is, in reality, a complementary strategy: while the affected collateral (WETH) requires stability for an orderly socialization of losses, the debt and exit asset (USDC) needs high rates that attract fresh liquidity and keep withdrawals functioning.
In any case, the success of this implementation will determine whether Aave can continue operating as resilient cryptoasset infrastructure in the face of systemic risk events in the DeFi sector.