Restaking Explained: The Complete Beginner’s Guide

Restaking started as a single idea inside Ethereum’s staking economy: what if the ETH already locked up to secure the base chain could be reused to secure other things too? A few years later, that idea has grown into a multi-billion dollar sub-industry with its own token economics, its own governance fights, a nine-figure exploit, and a regulatory question or two still unresolved.

This guide breaks restaking down from first principles: what it is, why it exists, how the mechanics actually work, who the major players are heading into the second half of 2026, where the yield really comes from, and what can go wrong when a security model built on reused trust meets a bridge exploit or a slashing event.

What Is Restaking?

Staking is the process of locking up cryptocurrency, typically ETH on Ethereum, to help run and secure a proof-of-stake blockchain. Validators lock 32 ETH, or delegate a smaller amount to a pool, propose and attest to blocks, and earn a native yield in return, currently in the high twos percent range annually. Misbehave, go offline for long stretches, or get caught double-signing, and the protocol slashes a portion of that stake.

Restaking takes that same locked capital, or a token that represents it, and pledges it a second time to secure something else entirely: an oracle network, a cross-chain bridge, a data availability layer, a rollup’s sequencer set, or one of dozens of other services that need real economic security but do not want to bootstrap an entirely new validator set and token from scratch. In exchange for opting in, the restaker earns additional rewards. In exchange for those rewards, the restaker accepts additional slashing conditions defined by whatever new service they are now also securing.

The term was popularized by EigenLayer, which coined “restaking” as a category and remains the largest protocol in the space, though the idea has since spread to Symbiotic, Karak, SSV Network’s Based Applications, and even to Bitcoin via Babylon. All of them share the same core insight: pooled, already-proven security is a scarce resource, and reusing it is cheaper than reinventing it from zero.

Why Restaking Exists: Bootstrapping Security for New Protocols

Every new blockchain-adjacent service faces the same cold-start problem. An oracle network that feeds price data to lending markets needs enough economic stake behind it that corrupting the feed costs more than an attacker could profit from manipulating it. A new data availability layer needs its own validator set willing to get slashed for withholding data. A bridge needs guardians who lose real money if they sign a fraudulent withdrawal.

The traditional answer was to launch a new token, convince people to buy and stake it, and hope the resulting market capitalization was large enough to make attacks unprofitable. That approach is slow, expensive, and mercenary: cheap, low-conviction capital tends to chase the highest yield and can leave just as fast, which is exactly the wrong property for a security budget.

Restaking offers a shortcut. Ethereum already has tens of billions of dollars in ETH staked by validators who have demonstrated they are willing to run infrastructure and accept slashing risk. Instead of asking those validators, or new entrants, to buy a fresh token, a new service asks them to extend their existing stake and their existing operational setup to a new set of rules. The service gets a running start on credible, sizable economic security. The validator or delegator gets extra yield on capital that was otherwise just sitting there earning the base staking rate. That is the trade at the center of every restaking protocol, and why the category grew as fast as it did once EigenLayer’s mainnet opened in 2023 and 2024.

How Restaking Actually Works: Deposits, Operators, AVSs and Slashing

Every restaking system, regardless of which protocol runs it, breaks down into four moving parts.

• Depositors (restakers) lock ETH, a liquid staking token like stETH, or in some protocols any approved ERC-20 asset, into a restaking contract. On EigenLayer this can be native ETH already staked as a validator, or a liquid staking token deposited fresh; on Symbiotic, almost any ERC-20 collateral can be wrapped and restaked.
• Operators run the actual infrastructure: servers, node software, uptime monitoring. A depositor delegates restaked capital to an operator rather than running services personally, the same way a liquid staking user delegates to a validator operator instead of running hardware themselves.
• AVSs, short for Actively Validated Services (EigenLayer’s term; Symbiotic calls them networks, SSV calls its version Based Applications), are the protocols actually being secured: oracle networks that feed price data into lending markets, a category whose price-feed manipulation history is exactly the kind of attack restaking-backed security is meant to make more expensive, plus cross-chain bridges, data availability layers, and rollup sequencer sets. Operators opt in to run an AVS’s specific software alongside their normal validator duties, and each AVS sets its own slashing conditions, reward schedule, and risk profile.
• Slashing is where the extra yield gets paid for. If an operator misbehaves under an AVS’s rules, whether by signing something fraudulent, going offline when the AVS requires liveness, or otherwise breaking that AVS’s specific conditions, a portion of the restaked capital delegated to that operator can be destroyed. This is separate from, and additional to, ordinary Ethereum consensus-layer slashing. A restaker who opts into five AVSs has, in effect, five new ways to lose money that did not exist when they were just staking ETH natively.

EigenLayer turned this from a whitepaper into a running system in stages. Its AVS marketplace opened on mainnet through 2023 and into 2024, but slashing itself, the sharpest edge of the whole model, did not go live until 17 April 2025, nearly two years later, reflecting how carefully the team tried to de-risk the rollout before making the losses real.

EigenLayer to EigenCloud: The Pioneer’s Evolution

EigenLayer remains the largest and most closely watched restaking protocol, and its own path says a lot about where the category is headed. Founded by Sreeram Kannan, it built EigenDA, its own data availability AVS, which became one of the first AVSs to go live on Ethereum mainnet back in April 2024 and has remained among the most widely used since. By mid-2026, the project counted well over a hundred AVSs in various stages of development, several dozen live on mainnet, and thousands of registered operators and depositor addresses.

Its total value locked tells a more turbulent story. EigenLayer’s TVL peaked near $19.7 billion in early 2026 before a stretch of LRT depegs, the Kelp DAO incident described below, and a broader cooldown in restaking speculation pulled it down substantially. DefiLlama showed it around $8.9 billion in March 2026, and estimates through the middle of the year have ranged anywhere from roughly $5 billion to the low teens of billions depending on the tracker and the week, a reminder that TVL figures in this sector are measured inconsistently across platforms and should be read as an order of magnitude rather than a precise number.

The protocol also rebranded partway through this period, from EigenLayer to EigenCloud, alongside a strategic pivot beyond pure restaking. EigenCloud now positions itself as a verifiable-compute platform, with products like EigenCompute and EigenVerify aimed at giving AI agents and off-chain applications cryptographic proof that their outputs were computed correctly. Kannan described the ambition on the Bankless podcast: “Anything you can program on the cloud, you should be able to program on Eigencloud with crypto-grade verifiability. If you do it on this, you don’t need to trust anybody.” The pivot has real financial backing behind it: a16z crypto, which had already led EigenLayer’s $100 million Series B in February 2024, purchased a further $70 million of EIGEN tokens directly in June 2025 specifically to back the EigenCloud launch.

On the token side, EIGEN has traded far below its all-time high throughout 2026. CoinGecko puts it around $0.23, with a market capitalization near $174 million as of mid-July 2026, down roughly 96% from its December 2024 peak of $5.65. A December 2025 governance proposal, ELIP-012, tried to address the gap between token price and protocol usage by building a new emissions controller that steers EIGEN issuance toward “productive stake”, meaning capital actively securing live, fee-generating AVSs, rather than idle deposits, and by routing cloud-product fees toward a buyback mechanism intended to reduce circulating supply over time.

Liquid Restaking Tokens: Yield Without Locking Capital

Restaking directly through EigenLayer or a similar protocol locks up capital in a way that mirrors the illiquidity of ordinary staking: once deposited, it is not easily usable anywhere else. Liquid restaking tokens (LRTs) solve that the same way liquid staking tokens solved it for plain staking: deposit ETH or an LST into a vault, and receive a tradeable receipt token that represents the restaked position and can still be used as collateral, swapped, or deployed elsewhere in DeFi.

ether.fi’s eETH is the largest LRT by deposits, with Renzo’s ezETH, Kelp DAO’s rsETH and Puffer’s pufETH making up most of the rest of a segment that has held several billion dollars in combined TVL through 2026. Advertised yields across these products have generally landed in the high single digits to low double digits annually, a mix of Ethereum’s native staking return, EigenLayer or Symbiotic points and rewards, and each protocol’s own token incentives.

The convenience comes with a second, stacked layer of risk on top of ordinary restaking risk: the receipt token itself can lose its peg if the market loses confidence in the vault, the bridge connecting it across chains, or the operators it delegates to. Renzo’s ezETH briefly depegged in April 2024 during a token-generation-event rush; Kelp DAO’s rsETH suffered a far more serious incident two years later, covered in detail below. Both cases point to the same lesson: an LRT is only as trustworthy as its weakest technical link, whether that is a smart contract, an oracle, or a cross-chain bridge, and any DeFi protocol that accepts an LRT as collateral inherits that weakest link whether it prices the risk in or not.

The Restaking Landscape in 2026

EigenLayer, now EigenCloud, may be the largest restaking platform, but it is no longer the only one. Three credible alternatives, plus an entirely separate Bitcoin-native branch, have emerged since 2025.

Symbiotic launched its mainnet on 28 January 2025 with a notably more permissive design: rather than restricting collateral to ETH and its liquid staking derivatives, it allows almost any ERC-20 token to be wrapped and restaked, and slashing was live from day one instead of being phased in over two years the way EigenLayer’s was. Disputes over whether a slashing condition actually triggered get resolved by an external “resolver”, which can be a decentralized oracle, an arbitration system, or a permissioned committee depending on how a given network is configured. Pantera Capital led a $29 million Series A round into the team behind Symbiotic in April 2025 alongside Coinbase Ventures, bringing its total funding to $34.8 million, according to CoinDesk. Despite roughly two years of mainnet operation, Symbiotic still had no public token as of mid-2026.

Karak is the third major entrant, using a broadly similar operator-and-AVS model but with a smaller footprint by most measures of deposited value, and with the public listing status of its own token still unclear as of this writing.

SSV Network approaches the same problem from a different angle. It is primarily a distributed validator technology (DVT) provider, splitting a single Ethereum validator key into shares held by multiple non-colluding operators, so that no single operator ever holds a complete key. In January 2025, SSV Labs layered a restaking-style model on top of that DVT base, called Based Applications: validators opt in through a separate participation key while the underlying 32 ETH principal stays unslashable, and only additional, voluntarily delegated capital is put at risk. SSV Labs founder Alon Muroch called it, in comments reported by Cointelegraph, the network’s “biggest, most ambitious project”, saying it “will profoundly change the restaking market.”

Comparing these systems on a like-for-like basis is genuinely difficult, since trackers disagree on methodology and the underlying figures move quickly. The table below should be read as a directional snapshot rather than a precise ranking.

Beyond Ethereum: Babylon and Bitcoin Restaking

Restaking is not exclusively an Ethereum phenomenon. Babylon extends the same basic idea, reusing already-committed capital to secure other systems, to Bitcoin, historically the hardest major asset to put to work without first wrapping it in a custodial or bridged form.

Babylon’s core mechanism uses Bitcoin’s own native timelock scripting to let BTC holders lock coins directly on the Bitcoin blockchain, with no wrapped token and no cross-chain bridge involved. That locked BTC then backs the security of proof-of-stake chains that opt in as what Babylon calls Bitcoin Supercharged Networks. If a validator misbehaves on one of those chains, the protocol can enforce a slashing path through the Bitcoin timelock itself.

Babylon’s Genesis mainnet and its BABY token launched together on 10 April 2025, with an airdrop of 600 million BABY, six percent of total supply, split across early stakers, staking-reward bonuses, Pioneer Pass NFT holders and open-source contributors. By mid-2026, Babylon’s own dashboard reported more than 56,800 BTC staked, worth roughly $5.6 billion, which the project describes as the largest Bitcoin staking system by that measure.

The BABY token itself has traded far below what that headline TVL figure might suggest. CoinGecko puts it around $0.013, down more than 90% from an April 2025 all-time high near $0.166, leaving a market capitalization around $51 million against several billion dollars of staked BTC, a valuation gap similar to the one EIGEN shows against EigenCloud’s TVL. A planned integration with Aave V4 would let natively staked BTC serve directly as lending collateral, though the timeline depends on Aave V4’s own rollout and a security review of the connecting infrastructure.

SSV, Obol and Distributed Validator Technology

Distributed validator technology (DVT) is a close cousin of restaking rather than restaking itself, and the two get discussed together because they solve adjacent problems for the same validator set.

Where restaking asks what else a staked deposit can secure, DVT asks how to stop a single validator key, and the single machine or operator holding it, from being one mistake away from a slashing event or an outage. SSV Network and Obol Network are the two leading answers. SSV splits a validator’s signing key into shares distributed across four or more independent operators using threshold cryptography, so the validator only acts when enough of those operators agree and no single operator ever holds the whole key. SSV describes itself as the largest Ethereum DVT provider by several measures, including total ETH secured, validator count, and operator count, in the multiple millions of ETH and thousands of operators, though independent trackers have at times shown meaningfully different numbers for all three, a reminder of how noisy public data can get across this sector.

Obol Network takes a similar approach with its Charon middleware, which sits between the validator client and the beacon node to coordinate a cluster-based consensus. Obol’s own token has fallen dramatically in price since its 2025 generation event even as Charon continues to run real validators inside Lido’s Simple DVT module, a stark divergence between actual protocol adoption and token-market sentiment. Lido’s own published figures show meaningful real-world DVT adoption through that module, which uses both Obol and SSV clusters across several hundred operators and thousands of validators combined, still a small share of Lido’s overall validator set but a concrete sign that DVT has moved well past the experimental stage.

Where SSV connects most directly to restaking is through its own 2025 pivot, Based Applications, described above, which layers an EigenLayer-style opt-in security marketplace on top of its existing DVT infrastructure rather than treating DVT and restaking as separate product lines.

The Risk Vitalik Warned About, and When It Became Real

Restaking’s core pitch, that idle economic security is wasteful and reusing it is efficient, has an equally serious counter-argument, and it did not come from a skeptic on the sidelines. Ethereum co-founder Vitalik Buterin raised it directly in a widely read blog post titled “Don’t overload Ethereum’s consensus”, published in May 2023 just as EigenLayer’s ideas were gaining traction. His argument: “Any expansion of the ‘duties’ of Ethereum’s consensus increases the costs, complexities and risks of running a validator.” He went further, warning about the social and political risk of restaking’s most aggressive use cases, the ones that ask Ethereum’s own validators to enforce rules for outside systems: “Blockchain communities’ social consensus is a fragile thing… each such extension makes the core itself more fragile.” The underlying fear is that if a restaking-secured protocol grows large enough, its failure could pressure the Ethereum community into an emergency fork to bail it out, exactly the kind of political entanglement proof-of-stake consensus is supposed to avoid.

Sreeram Kannan, EigenLayer’s founder, pushed back directly in a CoinDesk interview a few months later, arguing the risk was overstated relative to an alternative Ethereum users already tolerate: “Anything that restaking can do, already liquid staking can do… I view restaking as a lesser risk than liquid staking.” He also said he wanted the ecosystem to avoid assuming any single restaked protocol was “too big to fail” in a way that would force Ethereum to bail it out, rather than simply letting it fail and, in his words, “fork around” the failure.

That debate stopped being theoretical on 19 April 2026. Attackers exploited a cross-chain bridge used by Kelp DAO’s liquid restaking token, minting roughly 116,500 to 117,000 rsETH out of thin air, worth about $292 million, without ever touching Kelp’s actual deposits, according to CoinDesk’s reporting. The attackers then deposited the fraudulently minted rsETH as collateral on Aave and borrowed against it, leaving Aave with roughly $196 million in bad debt once the fraud was discovered. Aave’s total value locked fell from about $26.4 billion to roughly $20 billion over a single weekend, and the AAVE token dropped about 16% to around $92.

The recovery that followed doubled as a real-world test of the “too big to fail” question Kannan had raised years earlier. Aave founder Stani Kulechov posted through the crisis, quoted by CoinDesk: “rsETH has been frozen on Aave V3 and V4, the asset does not have any borrowing power as a measure due to KelpDAO bridge exploit that happened outside of Aave,” adding, “Aave is my life’s work and we’re working nonstop to find the best possible outcome for users”, and personally pledging 5,000 ETH toward the shortfall. A coalition that reportedly included Lido, ether.fi and Consensys joined the recovery effort. By around June 2026, rsETH was fully backed again, with Kelp migrating its bridge from LayerZero’s OFT standard to Chainlink’s CCIP and adding a requirement for four independent attestors and 64 block confirmations before minting; LayerZero, for its part, publicly disputed responsibility, characterizing the exploit as a consequence of Kelp’s own bridge configuration rather than a flaw in the base protocol.

The episode is a clean illustration of Vitalik’s original warning in a slightly different form than he described it: the systemic risk did not come from Ethereum’s consensus being overloaded directly, but from a restaking derivative being treated by another protocol as safe, fungible collateral without fully pricing in the risk of the bridge underneath it.

Where Restaking Yield Actually Comes From

Restaking dashboards advertise headline annual yields that can look dramatically higher than Ethereum’s roughly 2.7% to 2.8% native staking return, but that extra yield is not free money. Understanding where it actually originates matters for judging whether it is durable.

In practice, restaking yield is a blend of up to four distinct sources.

• Native staking yield. Restaked ETH is still staked ETH, earning Ethereum’s base return plus a modest boost from MEV, typically another half a percentage point to a full point.
• AVS-specific rewards. An AVS that wants operators and restakers to secure it pays for that security, usually in its own token, and that payment can be generous early on while a project bootstraps adoption and much thinner later.
• Points. Non-transferable, off-chain accounting units handed out with the implicit promise of a future token airdrop, exactly how EigenLayer, most LRTs, and SSV all built their earliest user bases before any of their tokens existed.
• Token emissions. Newly created supply distributed to stakers, which increases circulating supply over time and can dilute a token’s price even as the nominal yield looks attractive.

The gap between the first two categories, real yield paid for by real usage, and the last two, points and emissions, which are a promise and a dilution respectively, is the central question in restaking economics right now. EigenLayer’s own ELIP-012 governance change is a direct response to it, explicitly trying to redirect EIGEN issuance toward AVSs that are actually generating fee revenue rather than idle deposits. SSV Network made a similar move with its cSSV fee model, shifting rewards toward a share of real ETH-denominated staking fees rather than pure SSV token emissions. Whether either change closes the gap between headline yield and durable, revenue-backed yield is still an open question as of mid-2026; we cover that shift, protocol by protocol, in far more depth in our dedicated piece on restaking yield beyond the points era.

Regulation: Where the SEC Draws the Line

Restaking sits in a regulatory position that is more settled than it looks at first glance, but not fully settled either. The U.S. Securities and Exchange Commission has addressed plain staking directly: a Corporation Finance staff statement on 29 May 2025 concluded that protocol staking, whether solo, delegated, or run through a custodian, does not by itself involve a securities transaction, a position echoed in a companion statement from Commissioner Hester Peirce titled “Providing Security is not a ‘Security’”. A follow-up release on 5 August 2025 extended that same shelter to liquid staking and to staking receipt tokens such as stETH.

Restaking and liquid restaking tokens were not the explicit subject of either statement, and as of mid-2026 the SEC has not issued dedicated guidance addressing AVS reward tokens, LRT rehypothecation, or restaking-specific slashing risk. A broader interpretive release the agency published in March 2026, in which Chair Paul Atkins declared that “most crypto assets are not themselves securities” and that “this is what regulatory agencies are supposed to do: draw clear lines in clear terms”, leans in restaking’s favor by treating staking rewards generally as compensation for a service rather than a profit share, provided the platform exercises no discretion over the return. Whether that logic extends cleanly to an LRT that actively allocates deposits across multiple AVSs, each with a different risk and reward profile chosen by the protocol rather than the end user, is exactly the kind of question that has not yet been tested.

The legislative vehicle that could eventually settle questions like this at the statutory level, rather than through non-binding staff guidance a future commission could withdraw, is the CLARITY Act, which would formally divide digital asset oversight between the SEC and the CFTC. It passed the House in July 2025 but remained stalled in the Senate as of mid-July 2026, short of the 60 votes needed to clear a filibuster and held up by disputes unrelated to restaking itself: an ethics fight over federal officials’ personal crypto holdings, a law-enforcement provision that prosecutors argue could hamper criminal investigations, and a disagreement over whether stablecoin yield survives the final text. The Senate returned from recess on 13 July 2026 with roughly three working weeks left before its August recess, the window most analysts consider the last realistic chance for passage this year, and prediction markets tracked by Yahoo Finance had the odds of passage this year hovering in the mid-40s to mid-50s percent range as of this writing. We cover the bill’s mechanics and its path through Congress in detail in our explainer on the CLARITY Act.

Restaking vs Liquid Staking vs Solo Staking

With solo staking, liquid staking, and restaking now all live and competing for the same underlying ETH, the practical question for most holders is which one actually fits their goals.

Solo staking means running your own validator: locking 32 ETH, operating the client software, and keeping it online, in exchange for the full native yield with no intermediary and no smart-contract risk beyond Ethereum’s own protocol. It is the most decentralization-friendly option and the most technically demanding; we walk through the actual setup in our guide to running your own validator.

Liquid staking removes the technical burden and the 32 ETH minimum by pooling deposits and issuing a tradeable token like stETH in return, at the cost of delegating trust to whichever operator set the protocol uses and, at scale, raising centralization concerns of its own.

Restaking adds a further layer on top of either approach: the same capital, or its liquid representative, gets pledged again to AVSs for additional yield and additional risk. It does not replace solo or liquid staking, it extends whichever one a user has already chosen.

None of the three is strictly better than the others; they trade off decentralization, convenience, liquidity, and risk differently enough that the right answer depends on how much complexity, and how much additional slashing exposure, a given holder is willing to accept in exchange for extra yield.

Auditing the Stack: Security Review in a Reused-Trust System

Every layer added to restaking’s stack, the base protocol, the operator software, each individual AVS, and any LRT wrapped around the whole thing, is a separate piece of code that can contain a bug, and ideally a separate set of smart contracts that gets independently reviewed before it holds billions of dollars. That stacking is exactly what makes security review in this sector unusually hard: an audit of EigenLayer’s core contracts says nothing about the security of an individual AVS built on top of it, and an audit of an LRT’s vault contract says nothing about the cross-chain bridge it depends on, which is precisely the gap the Kelp DAO incident fell through.

Crypto’s leading smart-contract auditors, including firms like CertiK, have expanded their practices to cover exactly this kind of layered, cross-protocol risk, though the industry’s own record shows a clean audit is not a guarantee against a subsequent exploit; we cover that tension, and CertiK’s own scandals alongside its market position, in our explainer on CertiK. For restaking specifically, the practical implication is that a protocol’s own audit history covers only the layer that protocol actually wrote. Anyone allocating meaningful capital into an AVS or an LRT should assume that “audited” describes one link in a much longer chain, not the chain as a whole.

What’s Next for Restaking

Restaking’s next phase looks less like pure yield farming and more like infrastructure consolidation. EigenCloud’s pivot toward verifiable compute, aimed at giving AI agents and off-chain applications cryptographic proof of correct execution, is the clearest example: it reframes restaked ETH as collateral backing a computation-verification business rather than just a security budget for oracles and bridges. SSV’s cSSV fee model and Babylon’s push toward native BTC as DeFi collateral point in a similar direction, an attempt to tie token value to metered, real usage instead of speculative points programs.

Whether any of this closes the gap between token price and protocol scale that has defined the sector through 2026, EIGEN trading at a small fraction of its all-time high despite billions of dollars in TVL, SSV similarly depressed despite securing billions in validator capital, BABY worth tens of millions against several billion dollars in staked Bitcoin, is the question that will determine whether restaking becomes a durable piece of crypto’s financial plumbing or a cautionary tale about reusing trust one layer too many. The Kelp DAO incident showed what happens when that reuse goes wrong at the DeFi-integration layer. ELIP-012 and its counterparts at SSV and elsewhere are, in effect, an attempt to make sure the next multi-billion-dollar failure mode is a governance and incentive-design problem the industry can see coming, rather than a $292 million surprise.

Frequently Asked Questions

What is restaking in simple terms?

Restaking means taking crypto that is already staked to secure a base blockchain, most commonly ETH on Ethereum, and pledging it a second time to help secure a separate protocol, such as an oracle network, a bridge, or a data availability layer. In return, the depositor earns additional rewards on top of ordinary staking yield, and in exchange accepts additional slashing conditions defined by whatever new service they are securing.

What is the difference between staking, liquid staking and restaking?

Staking locks ETH directly to help run Ethereum itself and earns the network’s native yield. Liquid staking pools deposits through a protocol like Lido or Rocket Pool and issues a tradeable token, such as stETH, representing the staked position, so the holder keeps liquidity while still earning that native yield. Restaking takes either a direct stake or a liquid staking token and pledges it again, through a protocol like EigenLayer, Symbiotic or SSV, to secure additional services in exchange for extra rewards and extra risk.

Is restaking safe, and what happens if I get slashed?

Restaking carries more risk than plain staking, not less, because each additional service a validator or operator opts into comes with its own slashing conditions on top of Ethereum’s normal consensus rules. If an operator misbehaves under a specific AVS’s rules, a portion of the capital delegated to that operator for that AVS can be destroyed, separate from any base-layer staking slashing. Liquid restaking tokens add a further layer of risk from the bridges and vaults that wrap the underlying position, as the April 2026 Kelp DAO exploit demonstrated.

What is EigenLayer, and how is it different from Symbiotic or Karak?

EigenLayer, rebranded EigenCloud, is the largest and oldest restaking protocol, built around ETH and liquid staking tokens with a phased rollout that only enabled live slashing in April 2025. Symbiotic, launched in January 2025, takes a more permissive approach, accepting almost any ERC-20 as collateral and enabling slashing from day one, with disputes settled by an external resolver. Karak is a third, smaller competitor using a broadly similar model. All three compete for the same restaked capital and operator base, though EigenLayer/EigenCloud remains substantially larger by most measures of total value locked.

Can Bitcoin be restaked too?

Yes. Babylon lets Bitcoin holders lock BTC directly using Bitcoin’s own timelock scripting, without wrapping the coin or bridging it to another chain, and uses that locked BTC to help secure proof-of-stake networks that opt in. Babylon’s mainnet and BABY token launched in April 2025, and by mid-2026 the protocol reported more than 56,000 BTC staked, worth several billion dollars, making it the largest Bitcoin-specific staking system by that measure, even though the BABY token itself trades at a market capitalization far smaller than the value it secures.

Written by the HOGE Wire research desk.