LlamaRisk published an Aave Request for Comments on 10 November 2026 that would narrow the role of Aave V3’s Ethereum Core market. The proposal covers rsETH, osETH, LBTC, tBTC, rETH and several lower-exposure reserves. Its 9 November snapshot attributes $1.33 billion of collateral and $1.03 billion of debt to the assets in scope.

This is a governance proposal, not a completed market change. The published next step says that, after approval, the initial measures would be submitted as an Aave Improvement Proposal. NetNapz has not verified an approved vote, an executed payload or a finished migration.

What the initial measures would change

The proposal would freeze each in-scope reserve that is not already frozen. A freeze would stop new supply and new borrowing against that collateral and set its loan-to-value ratio to zero, including in relevant efficiency-mode categories. Existing positions would remain open. Borrowers could repay, withdraw or migrate rather than being closed automatically by the initial measure.

The distinction matters because rsETH and tBTC were already frozen when the snapshot was taken. For those two reserves, the proposal does not describe a new freeze. Other listed collateral, including osETH, LBTC and rETH, would receive new restrictions if governance approves and executes the plan.

Borrowable frozen reserves with more than $20,000 of outstanding debt would also move to a 99% reserve factor. The forum specification includes CRV, tBTC, rETH, crvUSD, FBTC, UNI and mUSD. That setting would direct most remaining borrower interest to the DAO rather than suppliers; it does not erase the debt or guarantee that positions migrate.

Most exposure sits in leveraged Ether loops

The proposal’s largest positions are rsETH and osETH. Together, the forum snapshot says they back about $862 million of WETH debt, largely through highly leveraged ETH-correlated loops. LBTC, tBTC and rETH account for a further $160 million of debt, mostly stablecoins, spread across more addresses and at lower leverage.

That concentration shapes the proposed exit route. Aave would add retained collateral such as wstETH, weETH and cbETH to selected efficiency-mode categories so some borrowers could swap collateral without first closing the whole position. The proposal says issuers and large borrowers would be approached directly, while positions would also be allowed to unwind naturally.

Later controls would require separate action

The proposal authorises ranges for follow-up controls only if exposure does not decline. For LBTC, tBTC and rETH, liquidation-threshold reductions could proceed in steps of up to two percentage points, subject to limits on the amount made liquidatable. For rsETH and osETH loops, a WETH borrow-rate increase of up to 100 basis points is described as a final measure, alongside targeted rebates for borrowers who do not hold the affected collateral.

Those later measures are neither immediate nor automatic. The forum says they would require Risk Steward action or separate direct-to-AIP proposals, would be announced in advance, and should be sized to avoid abrupt liquidations. The proposal therefore establishes a possible control range rather than proof that any specific rate or liquidation threshold will change.

NetNapz assessment: track governance, execution and migration separately

NetNapz assessment: the practical risk signal is the sequence, not the headline collateral total alone. A forum proposal shows the intended policy. A successful vote would show governance approval. An executed payload would show that reserve parameters changed. Later exposure reports would show whether borrowers actually migrated without destabilising liquidations.

The proposal also marks a structural division between Aave V3 Core and Aave V4. V3 Core would concentrate on established, liquid collateral, while V4’s hub-and-spoke model is presented as the venue for a broader set of risks with asset-specific caps and pricing. That design claim should still be tested against live configurations, liquidity and borrower behaviour once deployments occur.

Verification scope: this report uses the dated LlamaRisk governance proposal and its 9 November exposure snapshot. It does not assert that the ARFC has passed, that an AIP has executed, that any asset price will fall, or that the quoted collateral and debt totals remain current after the snapshot.