Here's a thought experiment.
Ethereum has spent years accumulating billions of dollars in staked ETH, with validators putting up collateral to secure the network. That security is expensive to build. It took years. And once a validator's ETH is staked to secure Ethereum, it just sits there, doing one job.
What if it could do two?
That's the question EigenLayer answered. And the answer is reshaping how crypto networks get built. But to understand why it matters, we need to start with how Ethereum security actually works.
Welcome to Day 58 of 60-Days in Web3 series. If you want to follow along as I keep learning, building, and occasionally changing my mind about Web3, you can find the rest of this series on X, on Medium, on Future, and you can join the Web3ForHumans Telegram community to discuss these topics in plain language.
How Ethereum security works
When you stake ETH on Ethereum, you're putting up collateral and running a validator node. Your job is to honestly process and confirm transactions. If you behave honestly, you earn rewards. If you try to cheat, your stake gets slashed, meaning you lose a portion of it as punishment.
This slashing mechanism is what makes the network trustworthy. Attackers would need to control 33% of staked ETH to meaningfully disrupt the network. At current staking levels, that's tens of billions of dollars of exposure. It's not worth it.
That security is Ethereum's moat. Building it took years of validator participation and accumulated trust.
Now what if a new protocol, a data availability layer, an oracle network, a bridge, wanted that same level of security?
Normally, they'd have to build it from scratch. Attract stakers. Issue their own token. Build their own slashing conditions. Hope validators show up.
EigenLayer said: what if they didn't have to?
What EigenLayer actually is
EigenLayer is a protocol that lets Ethereum validators opt in to securing additional services, called Actively Validated Services (AVSes), using their already-staked ETH.
The staker still secures Ethereum. But they also, voluntarily, extend their security to another network or service. In exchange, they earn additional yield from that AVS.
The AVS gets Ethereum-grade security without having to bootstrap its own validator set. The staker earns more. The broader ecosystem gets more secure infrastructure, faster.
This concept is called restaking. And it comes in two forms.
The mechanics
Native restaking. If you're running a full Ethereum validator, you can point your withdrawal credentials to EigenLayer's contracts. Your staked ETH now simultaneously secures Ethereum and any AVSes you've opted into.
Liquid restaking. If you hold staked ETH derivatives like stETH from Lido or rETH from Rocket Pool, you can deposit those into EigenLayer directly. You don't need to run your own validator.
Both approaches let validators and token holders extend their security work to additional services and get paid for it.
The services on the receiving end of that security are what EigenLayer calls AVSes, and that's where it gets interesting.
What AVSes are and why they need this
An AVS is any service that needs decentralized validation. Things like:
- Data availability layers, services that ensure transaction data is accessible and verifiable, essential for rollups. Building trust for these from scratch is slow and expensive.
- Decentralized oracles, networks that need honest nodes reporting real-world data to blockchains. The more economic weight behind a node, the more trustworthy the data.
- Cross-chain bridges, which have historically been among the most exploited pieces of Web3 infrastructure. Restaked security adds a meaningful economic deterrent.
- Sequencer networks, since L2s often rely on a single centralized sequencer to order transactions. Decentralized sequencer networks need validators, and restaking gives them a path to get them.
EigenLayer's first major AVS is EigenDA, a data availability layer that already serves several major rollups.
Once restaking started gaining traction, a new category of protocols emerged to sit on top of it.
Liquid restaking tokens (LRTs), the next layer
As restaking grew, a new category emerged: liquid restaking protocols.
The idea is straightforward. You restake your ETH through EigenLayer, but your capital is locked and illiquid. Liquid restaking protocols like Ether.fi, Renzo, Kelp DAO, and Puffer Finance take your ETH, restake it on your behalf across AVSes, and give you a liquid token, an LRT, in return.
That LRT can then be used in DeFi. Lend it, provide liquidity with it, use it as collateral.
Your ETH is earning Ethereum staking rewards, EigenLayer restaking rewards, and potentially DeFi yield, all simultaneously. The composability is powerful.
It's also where the risk starts to build up.
The risks, and they're real
Slashing cascades. If a validator behaves maliciously on an AVS, or if an AVS has a bug in its slashing conditions, the validator's restaked ETH could be slashed across multiple services simultaneously. The exposure is compounded, not isolated.
AVS quality variance. Not all AVSes are equal. Opting into a poorly audited AVS is opting into risk. Stakers have to evaluate each service individually, which most retail participants are not equipped to do.
Correlated risk concentration. If a large liquid restaking protocol controls a huge share of restaked ETH and encounters a problem, it affects every AVS that protocol's validators were securing. The interconnection creates systemic risk.
Points farming distortion. During EigenLayer's launch period, protocols offered points, pre-token incentives, to attract restakers. This created massive capital flows chasing points rather than evaluating risk. Misaligned incentives.
The social consensus problem. Vitalik Buterin himself wrote about this: the concern that EigenLayer could become so large that social pressure would emerge to bail out AVSes that fail, effectively turning Ethereum's social consensus into collateral for third-party services. The community is watching this closely.
What's happened since launch
EigenLayer opened restaking to the public in 2024. By early 2025, it had accumulated tens of billions in restaked ETH, making it one of the largest protocols in DeFi by TVL in a remarkably short period.
The liquid restaking token market exploded alongside it. Ether.fi became one of the fastest-growing DeFi protocols ever. Most major L2s announced or completed integration with EigenDA.
EIGEN, EigenLayer's native token, launched with a novel design: it's meant specifically for intersubjective tasks, slashing conditions that can't be programmatically verified on-chain but require human judgment to resolve. That's another piece of the emerging security economy worth watching.
The bigger picture
Restaking represents a shift from seeing Ethereum as a single, isolated network to seeing it as a security marketplace, one where economic weight can be directed at different services, composable and programmable.
That's a genuinely different idea of what a blockchain is. Not a standalone ledger, but the foundation layer of a security economy.
Every new protocol that opts into EigenLayer instead of bootstrapping its own validator set is a vote for this model. And given how fast the TVL and AVS count grew, a lot of protocols are casting that vote.
Whether the risk compounds faster than the ecosystem can manage it is the open question. The answer will play out over the next few years in real time, in the most adversarial financial environment in the world.
Tomorrow, we'll look at something different: the meta-story of what actually happened across 58 days of writing about Web3, the patterns, the surprises, and what the data says about this space.
Resources to go deeper
The EigenLayer whitepaper is the primary reference for how restaking and AVSes work at a protocol level.
The EigenLayer docs cover native restaking, liquid restaking, and the AVS operator framework in practical detail.
For a critical perspective, Vitalik Buterin's post Don't overload Ethereum's consensus lays out the social consensus risk clearly and is essential reading alongside the bullish case.
EigenDA's documentation explains how the first major AVS works and which rollups are already using it.
For the liquid restaking layer, Ether.fi, Renzo, and Kelp DAO each have docs that explain how LRTs work in practice, and comparing them is a useful exercise.
If you want to follow along as I keep learning, building, and occasionally changing my mind about Web3, you can find the rest of this series on X, on Medium, on Future, and you can join the Web3ForHumans Telegram community to discuss these topics in plain language.
Top comments (0)