How Does ether.fi Work?
ether.fi is the second-largest liquid restaking protocol globally with $5.77B in deposits, offering eETH — a liquid token that simultaneously earns Ethereum staking rewards and EigenLayer restaking rewards — alongside ether.fi Cash, a Visa debit card product that allows users to spend DeFi yields with up to 3% cashback. Its C- grade reflects three novel mechanisms: EigenLayer restaking with socialized slashing across all eETH holders (live since April 2025), non-custodial ECIES/IPFS validator key management, and Borrow Mode (a DeFi CDP-backed Visa card). Elevated regulatory risk stems from the April 2025 neobank pivot and the Cash product's dependency on undisclosed banking partners and Visa network compliance. Despite ~$166M in annualized protocol revenue and 28 audit reports with Certora formal verification, the combination of restaking complexity and consumer financial product regulation places this above typical liquid staking protocols.
TVL
$5.2B
Sector
Restaking
Risk Grade
C-
Value Grade
B-
Core Mechanisms
2.1.1
eETH rebasing liquid staking token (LiquidityPool → eETH mint/redeem)
Standard Lido-pattern LST (since 2020). Users deposit ETH, receive eETH (rebasing). Balance formula: TotalPooledEth × (shares/totalShares). 90% of rewards to stakers, 5% to node operators, 5% to protocol treasury.
2.1.2
weETH non-rebasing ERC-4626 wrapper for DeFi compatibility
Standard wrapped LST pattern. weETH is used as collateral in Aave, Morpho, Pendle, Balancer, Uniswap V3. OFT adapter enables cross-chain deployments (Arbitrum, etc.).
2.2.1
NovelEigenLayer restaking with socialized AVS slashing across all eETH holders
Novel: EigenLayer restaking with socialized slashing is <3 years old (launched 2023-2024) and not replicated by 3+ major protocols. All eETH holders share proportional slashing losses from any participating AVS. Live slashing since April 2025.
2.3.1
NovelNon-custodial ECIES/IPFS validator key management (encrypted keys stored on IPFS)
Novel: ECIES public-key encryption of validator keys with IPFS storage and on-chain hash commitment. Operators decrypt via shared secret without ever holding plaintext keys. Not replicated by 3+ major protocols. DVT upgrade to SSV Network in progress.
4.1.3
EtherFiOracle permissioned committee for eETH exchange rate reporting
Permissioned committee oracle that aggregates beacon chain + EigenLayer rewards and updates TotalPooledEth periodically. Custom design but multi-party. Used for reward accounting, not real-time price feeds.
3.2.1
NovelBorrow Mode: ETH/eETH collateral → Aave USDC borrow → Visa card spend
Novel: DeFi CDP-backed Visa card spending. Users deposit collateral, protocol borrows USDC from Aave on their behalf, funds a Visa debit card. Not replicated by 3+ major protocols for 3+ years. Requires KYC and banking partner.
5.1.1
ether.fi Cash Visa debit card (Spend Mode + Borrow Mode, migrating to OP Mainnet)
Fiat payment rail product. 70,000+ active cards, 300,000+ users, ~$2M/day spend. Originally on Scroll, migrating to OP Mainnet (announced Feb 2026). Requires Visa network + card-issuing bank partnership.
1.3.1
Multi-strategy Liquid Vaults (ETH/BTC/USD) built on Veda infrastructure
Standard yield aggregator pattern (Yearn-style). Three automated strategy vaults deploy across Pendle, Aave, Morpho, Balancer, Convex, Aura, Uniswap V3. 2% annual management fee. US liquid reserve vault available.
6.1.1
ETHFI governance token (1B fixed supply, fully minted, DAO-governed)
Standard governance token. 1 billion fixed supply (fully minted at TGE March 2024). DAO governs protocol parameters, AVS selection, fee splits. Snapshot + on-chain governance at etherfi-dao.eth.
6.2.1
ETHFI buyback program funded from protocol revenue (staking fees + vault management fees + Cash interchange)
Governance-controlled buyback program. Revenue sources: 5% staking reward share, 2% annual vault management fee, Cash interchange/FX, swap fees. 30-day revenue ~$13.9M. Buybacks are discretionary, not on-chain enforced.
How the Pieces Interact
An AVS slashing event reduces eETH's TotalPooledEth, proportionally reducing all holders' balances simultaneously. A large slashing event could trigger mass exits, overwhelming the 7-day unbonding queue and causing eETH to trade at a discount on secondary markets.
During coordinated market stress, simultaneous withdrawal requests cannot be fulfilled in real-time due to EigenLayer's 7-day unbonding period. eETH/ETH peg may break as secondary market liquidity is exhausted, affecting all DeFi protocols using weETH as collateral.
A 30-40% ETH price drop triggers Aave to liquidate user collateral positions in Borrow Mode, resulting in partial or full loss of deposited ETH. Users spending via the Visa card may not be able to add collateral in time if card functionality is temporarily disrupted.
A delayed or incorrect oracle report could cause a stale eETH exchange rate to propagate to lending protocols using weETH as collateral, potentially triggering incorrect liquidations or allowing under-collateralized borrowing during the lag window.
If the undisclosed banking partner faces regulatory action or exits, Cash product becomes non-functional. The 70,000+ active card users with Borrow Mode positions must immediately unwind collateral, potentially during adverse market conditions without card access.
What Could Go Wrong
- EigenLayer restaking with socialized slashing: all eETH holders share proportional losses if an AVS is slashed. EigenLayer's live slashing system (since April 2025) makes this an active risk — a major AVS incident could reduce eETH's value for all holders simultaneously.
- eETH withdrawal queue pressure during market stress: EigenLayer's 7-day unbonding period means simultaneous exits cannot be fulfilled immediately. Under market stress, eETH could trade at a discount on secondary markets, creating collateral risk for the ~$5.77B of positions using weETH in Aave, Morpho, and Pendle.
- Borrow Mode liquidation risk: Cash users who deposit ETH/eETH as collateral for USDC Visa spending can be automatically liquidated via Aave if ETH prices fall 30-40% below their LTV thresholds — a standard market movement in crypto.
- Neobank regulatory surface: the April 2025 pivot and Visa card product require compliance with MSB regulations, KYC/AML requirements, and banking partner dependencies. The undisclosed card-issuing bank creates a single regulated counterparty whose exit would disable the Cash product for 70,000+ active cardholders.
AVS Mass Slashing Cascade → eETH Depeg → weETH Collateral Liquidations
ModerateTrigger: A major EigenLayer AVS (e.g., one securing >$200M in restaked ETH) is found misbehaving and EigenLayer governance votes to execute a slashing of 5-10% of the AVS's restaked stake within a 7-day period
- 1.EigenLayer governance approves slashing of a major AVS participating in ether.fi's restaking — EtherFiOracle committee updates TotalPooledEth downward, reducing eETH balances proportionally for all holders (~0.5-1% of eETH value lost if 10% of restaked ETH is slashed)
- 2.eETH exchange rate drops; news of slashing creates panic among holders — Mass withdrawal requests submitted to ether.fi LiquidityPool; requests enter the EigenLayer 7-day unbonding queue; immediate liquidity in the exit pool is rapidly exhausted
- 3.eETH trades at 3-8% discount on secondary markets (Curve, Uniswap V3) as exit queue backs up — weETH price feeds in Aave and Morpho detect the discount; protocols reduce weETH LTV ratios or trigger liquidations on positions below updated collateral thresholds
- 4.Cascading weETH liquidations in Aave/Morpho create additional eETH sell pressure — Secondary market liquidity for eETH/weETH is depleted; discount widens to 10%+; Borrow Mode users with ETH/eETH collateral face liquidations from the combined price impact of ETH volatility + eETH discount
- 5.Protocol TVL drops sharply as users withdraw staking positions and vault TVL declines — ETHFI price falls due to reduced protocol revenue and negative sentiment; ETHFI buyback program becomes less effective; some Pendle/Liquid vault positions suffer compounded losses from weETH depeg + underlying protocol exposure
Risk Profile at a Glance
Overall: C- (56/100)
Lower score = safer