LibraryScaling2023Design paperCorpus record
EigenLayer: The Restaking Collective
EigenLayer. Eigen Labs.
Eigen Labs' design for letting Ethereum stakers opt in to additional slashing conditions, so new services can rent economic security instead of bootstrapping a new token set from zero.
EigenLayer's paper proposes that ether already staked to secure Ethereum can be restaked, under extra slash conditions, to secure other services the staker opts into.
The five-minute read
The economic security already exists
Ethereum validators post collateral and can be slashed for breaking Ethereum's rules. The paper asks them to accept additional rules, for additional pay, with the same collateral.
Opt-in is the whole permission system
A staker chooses which services, called AVSs in the paper, to secure. There is no automatic claim on someone's stake by a service they did not join.
Slashing conditions are the product
Each service defines what misbehaviour looks like and how it is proved. A vague condition is an unenforceable loan of security. The paper's safety is only as clear as those definitions.
Operators and restakers can be different people
The capital and the machine that runs the extra software need not belong to the same party. Delegation is part of the design and part of the risk.
This is not a 2023 measurement of Ethereum
The paper is a mechanism. How much stake opted in, and into what, is a fact that moves. Do not cite the paper as a capacity figure.
One action, walked through
- A validator, or a holder of liquid staking tokens, opts into the restaking contract and names the services they will secure.
- An operator runs the service's software and is delegated that stake, if the two roles are split.
- The service pays a fee for the security it is renting.
- If the operator breaks a stated rule, anyone able to prove it submits the proof, and the contract slashes the opted-in stake.
- The same stake remains subject to Ethereum's own slashing. The two liabilities stack.
The argument, unpacked
Reusing collateral reuses risk
One pool of ether backstopping many services is capital-efficient and correlated. A bug, or a mass slash, hits the same economic base that secures the chain the services were meant to lean on. The paper has to be read as a concentration design, not only as a marketplace.
Subjective slashing is a governance leak
If misbehaviour cannot be proved on-chain and a committee decides instead, the staker has trusted that committee with the collateral. The paper's crypto-economic story is strongest when the fault is objective. Services with subjective faults have changed the trust model.
Delegation creates a new operator class
Restakers who do not run the software will pick operators on yield and reputation. That is a market, and it will concentrate. The design does not prevent a few operators from becoming the de facto security for many services.
What has to be true
- Slashing conditions are objective enough to encode, and the proof of a fault fits in the contract.
- Stakers understand they can lose ether for failures that are not Ethereum consensus failures.
- The extra yield compensates the extra risk. The paper does not guarantee that it does.
- Ethereum itself remains the base slash. Restaking does not replace attestation duties.
What happened after the paper
EigenLayer launched restaking and a set of services after the paper, with parameters, caps and operator sets that the 2023 text does not freeze. The durable idea is optional extra slash conditions on existing stake. The live book of risk is whatever stakers have actually opted into.
What to check before you use the idea
- What, exactly, gets slashed, and is that fault provable on-chain?
- Is the operator the same party as the staker?
- How many services does this collateral already backstop?
- Does the extra yield still exist after a realistic slash, or only in the advertisement?
Terms
- Restaking
- Putting stake that already secures a base chain under additional slashing rules.
- AVS
- An actively validated service: the extra system that rents the restaked security.
- Operator
- The party who runs the service software, often with stake delegated by someone else.
- Slashing condition
- The rule that defines a punishable fault. If it is vague, the security is vague.
The problem the paper names
A new network that needs a bond usually issues a new asset and hopes it becomes expensive to attack. EigenLayer's paper proposes that an existing staked position can take on extra duties, and extra penalties, if the holder opts in.
What the design proposes
- Restakers point stake at operators. Operators run services called AVSs in the paper's vocabulary.
- Each service defines its own slashing condition. Opting in is specific, not a blanket lien on all staked ether.
- The base staking protocol remains Ethereum's. EigenLayer is an additional opt-in layer.
How the mechanism is specified
- Security is pooled only to the extent holders accept the new penalty. Capital and risk both move.
- An operator who runs many services concentrates operational risk even if the stake is widely held.
- Withdrawal delays and veto processes, if any, are part of the risk and have to be read in the current docs as well as the paper.
What this page does not treat as proven
- Restaking does not make a service correct. It makes a defined failure expensive for those who opted in.
- The paper is not an audit of any particular AVS.
- We do not describe yields. A slashing design is not a savings product.
Why a venture studio still reads it
The question for a venture is whether it needs its own bond or whether it can state a slashing condition an existing staker could actually evaluate. If the condition cannot be evaluated, restaking is just leverage with extra steps.
This is Blockchain Lab's reading of a public design paper. It is not the paper, not a copy of it, and not an offer of tokens, equity, custody or a partnership. Later network behaviour can diverge from the text. Nothing here is investment, legal or technical advice.
Research status: Design paper. Last reviewed: 1 October 2026. This is a reading of a public paper, not investment, legal or security advice.