Free 30-min Web3 Consultation
Book Consultation
Smart Contract Security Audits
View Audit Services
Custom DeFi Protocol Development
Explore DeFi
Full-Stack Web3 dApp Development
View App Services
Free 30-min Web3 Consultation
Book Consultation
Smart Contract Security Audits
View Audit Services
Custom DeFi Protocol Development
Explore DeFi
Full-Stack Web3 dApp Development
View App Services
Free 30-min Web3 Consultation
Book Consultation
Smart Contract Security Audits
View Audit Services
Custom DeFi Protocol Development
Explore DeFi
Full-Stack Web3 dApp Development
View App Services
Free 30-min Web3 Consultation
Book Consultation
Smart Contract Security Audits
View Audit Services
Custom DeFi Protocol Development
Explore DeFi
Full-Stack Web3 dApp Development
View App Services
layer-2-wars-arbitrum-optimism-base-and-beyond
Blog

Why Shared Security Models Will Centralize Power

An analysis of how the Superchain thesis and security-as-a-service models, championed by Optimism and others, will inevitably concentrate validator control, sequencer power, and governance influence into a few core entities, undermining decentralization.

introduction
THE CENTRALIZATION TRAP

Introduction

Shared security models, while solving for validator decentralization, inherently consolidate economic and governance power.

Shared security centralizes capital. Protocols like EigenLayer and Babylon aggregate restaked assets, creating a single point of economic failure and systemic risk that dwarfs individual chain vulnerabilities.

Governance power follows capital. The largest restakers (e.g., Lido, Coinbase) will dominate the slashing committees and governance votes of hundreds of actively validated services (AVS), creating a meta-governance cartel.

This creates protocol homogeneity. AVS builders optimize for the largest, lowest-cost security provider, leading to a monoculture of economic security that stifles innovation in consensus and reduces the attack surface diversity of the broader ecosystem.

thesis-statement
THE INCENTIVE FLOW

The Centralization Thesis

Shared security models create winner-take-all dynamics that concentrate power in a few dominant protocols.

Economic gravity favors incumbents. Protocols like EigenLayer and Babylon monetize security by attracting restaked capital, creating a liquidity moat. New entrants cannot compete with established Total Value Secured (TVS) figures, leading to market consolidation.

Validator centralization is inevitable. Capital follows the highest yield, which aggregates into the largest staking pools. This recreates the Lido problem at the infra layer, where a handful of node operators control the network's liveness.

Modular stacks create dependency. Rollups using Celestia for data or EigenDA for data availability delegate critical security. This creates systemic risk where a failure in the shared resource cascades across hundreds of chains.

Evidence: EigenLayer holds over $15B in restaked ETH, creating a security budget that dwarfs all new L1s combined. This capital concentration dictates which AVSs succeed.

market-context
THE INCENTIVE MISMATCH

The Current Landscape: A Rush to Franchise

Shared security models create a centralizing flywheel where economic power dictates technical control.

Shared security centralizes governance. Protocols like EigenLayer and Babylon sell pooled security as a commodity, but the capital providers who stake the most become the ultimate arbiters of network upgrades and slashing decisions.

Economic power dictates technical control. A franchise model emerges where the security provider's chain, like Cosmos Hub or Ethereum, becomes a de facto parent company, and the secured chains become franchisees with limited sovereignty.

The validator set ossifies. Major staking providers like Lido and Coinbase dominate the staking pools, creating a small, professionalized cartel that validates across hundreds of franchised chains, replicating the centralization of cloud providers.

Evidence: In Ethereum's restaking ecosystem, the top 5 node operators control over 60% of EigenLayer's TVL, creating a concentrated point of failure for dozens of actively validated services (AVS).

SHARED SECURITY MODELS

Control Matrix: Who Holds the Keys?

Comparing the power distribution and centralization vectors across major shared security implementations.

Control DimensionEigenLayer (Restaking)Cosmos Hub (Replicated Security)Polkadot (Parachains)Rollup-as-a-Service (RaaS)

Validator Set Control

Ethereum PoS Validators

Cosmos Hub Validators

Polkadot Relay Chain Validators

RaaS Provider (e.g., Conduit, Caldera)

Slashing Authority

EigenLayer AVS Operators

Cosmos Hub Validators

Polkadot Governance

RaaS Provider

Governance Upgrade Path

EigenLayer DAO + AVS-specific

Cosmos Hub Governance

Polkadot OpenGov + Parachain

RaaS Provider Admin Keys

Minimum Viable Trust Assumption

Honest majority of Ethereum stake

Honest majority of ATOM stake

Honest majority of DOT stake

Honesty of RaaS operator

Exit/Withdrawal Period

~7 days (Ethereum withdrawal queue)

21 days (Cosmos Hub unbonding)

28 days (Polkadot unbonding)

Instant (provider-dependent)

Economic Capture Risk

High (LST/ETH concentration)

High (ATOM concentration)

High (DOT concentration)

Absolute (single entity)

AVS/Parachain Count Limit

Theoretically unlimited

Practically limited by validator load

Capped by parachain slots (~100)

Provider capacity limit

deep-dive
THE INCENTIVE MISMATCH

The Slippery Slope: From Service to Sovereignty

Shared security models, like restaking and shared sequencers, create an economic incentive for providers to capture and centralize the very networks they serve.

Security-as-a-Service centralizes power. Providers like EigenLayer and Espresso Systems sell modular security, but their economic model rewards accumulating more stake and sequencer rights. This creates a winner-take-most market where capital efficiency dictates consolidation, not decentralization.

Sovereignty is the ultimate product. The endgame for a successful security provider is not being a vendor but becoming the chain. A dominant restaking pool or sequencer network has the capital and control to launch its own L2, directly competing with its customers like Arbitrum or Optimism.

Modular stacks re-centralize. The promise of modularity fragments execution but re-concentrates consensus and sequencing. A network like Celestia provides data availability, but the real power accrues to the few entities, like AltLayer or Caldera, that bundle the most attractive restaked security and sequencer services.

Evidence: Ethereum's proposer-builder separation (PBS) shows this dynamic. While intended to decentralize, it created a builder market dominated by a few like Flashbots, who now control transaction ordering. Shared sequencer networks will replicate this centralization at the L2 level.

counter-argument
THE CENTRALIZATION TRAP

The Rebuttal: "But It's More Secure!"

Shared security models like restaking and interchain security create systemic centralization by concentrating economic power and governance control.

Security is a political problem. Shared security models like EigenLayer restaking and Cosmos Interchain Security conflate economic security with political decentralization. Concentrating validation power in a few liquid staking tokens (LSTs) like Lido's stETH creates a single point of political failure for dozens of networks.

Economic gravity centralizes. Validators optimize for yield, not sovereignty. They will flock to the largest, most lucrative restaking pools, creating validator oligopolies. This mirrors the centralization seen in Ethereum's beacon chain, where Lido and Coinbase dominate stake.

Governance becomes extractive. The security-providing chain (e.g., Ethereum, Cosmos Hub) holds veto power over its clients. This creates a hub-and-spoke hierarchy where innovation in the spokes is subject to the hub's political whims, stifling permissionless experimentation.

Evidence: In Cosmos ICS, the ATOM token holds governance over consumer chains. In Ethereum's restaking future, a Lido DAO vote could dictate the slashing conditions for hundreds of actively validated services (AVS), centralizing critical protocol decisions.

case-study
WHY SHARED SECURITY MODELS WILL CENTRALIZE POWER

Case Studies in Concentrated Control

The promise of pooled security creates new, more opaque forms of systemic risk and control.

01

The EigenLayer Restaking Monopoly

EigenLayer's restaking model consolidates $18B+ in TVL under a single slashing contract. This creates a systemic risk hub where a bug or governance failure could cascade across dozens of actively validated services (AVSs). The economic gravity inherently centralizes power.

  • Single Point of Failure: Centralized slashing logic for all AVSs.
  • Capital Inefficiency: Capital is locked, not secured, creating artificial scarcity.
  • Opaque Risk Bundling: Stakers cannot accurately price the correlated failure risk of the AVSs they secure.
$18B+
TVL Locked
1
Slashing Contract
02

Cosmos Hub's Failed Interchain Security

Cosmos Hub's Interchain Security (ICS) has seen minimal adoption, with only a handful of consumer chains. The model fails because sovereignty is a primary value proposition for app-chains. The result is a security cartel where only the largest chains (like Cosmos Hub) can offer security, reinforcing a hub-and-spoke hierarchy.

  • Sovereignty Tax: Consumer chains cede economic and upgrade sovereignty.
  • Concentrated Validator Set: The same ~180 validators secure most of the Cosmos ecosystem, creating political centralization.
  • Economic Misalignment: Validators prioritize the hub's revenue over consumer chain health.
~180
Validators
<5
Active Consumers
03

Lido's Liquid Staking Governance Capture

Lido's $34B+ in staked ETH demonstrates how a shared utility becomes a governance bottleneck. The Lido DAO now holds veto power over the largest staking pool's technical upgrades and fee structure. This isn't just market share; it's protocol-level control over a core Ethereum primitive.

  • Veto Power: Lido DAO governs critical oracle and withdrawal credentials.
  • Barrier to Entry: Scale begets scale, stifling competition.
  • Centralized Oracle: The protocol relies on a 9-of-15 multisig for critical operations, a regression in trust assumptions.
$34B+
Staked ETH
9-of-15
Oracle Multisig
04

Celestia's Data Availability Cartels

Celestia's modular DA layer shifts power from block producers to data availability committees (DACs) and the highest bidders for blob space. This creates a capital-intensive oligopoly where only well-funded rollups can guarantee timely data posting, centralizing the rollup landscape itself.

  • Pay-to-Play Security: DA security is a function of capital, not decentralization.
  • Committee Risk: Reliance on small, permissioned DACs for high-throughput chains.
  • Vertical Integration: Large rollup teams will internalize DA sampling to reduce costs, creating walled gardens.
~100
Active Nodes
Oligopoly
Market Structure
05

The Interoperability Hub Trap

Protocols like LayerZero, Axelar, and Wormhole are becoming centralized message routers. Their security models (Oracle/Relayer networks, MPC committees) are controlled by the same small set of ~20-30 entities. This recreates the SWIFT problem: a few private companies control the plumbing for $100B+ in cross-chain value.

  • Trusted Committee: Security relies on a known, permissioned set of actors.
  • Protocol Lock-in: Applications design their architecture around a specific hub's features.
  • Censorship Vector: A small committee can theoretically freeze or censor message flows.
20-30
Core Entities
$100B+
Secured Value
06

The Re-Staking Liquidity Crisis

Shared security models like EigenLayer create a liquidity black hole. Capital locked for securing one service cannot be used elsewhere, creating artificial scarcity and driving yields down for all other DeFi. This centralizes economic activity around the dominant restaking platform, starving the broader ecosystem.

  • Capital Sterilization: $18B+ in ETH is removed from productive DeFi use.
  • Yield Compression: Redirects yield from DEXs/Lending to the restaking platform.
  • Systemic Illiquidity: A mass unstaking event could freeze multiple AVSs and DeFi simultaneously.
$18B+
Sterilized Capital
High
Correlation Risk
future-outlook
THE POWER DYNAMIC

The Inevitable Endgame: Cartels and Counter-Movements

Shared security models structurally incentivize the formation of dominant validator cartels, centralizing power they were designed to distribute.

Economic gravity centralizes staking. Validator sets in EigenLayer, Babylon, or Cosmos compete for restaking yield. The largest operators with the lowest operational costs and best MEV extraction (like Figment, Chorus One) capture market share, creating an oligopoly.

Cartels dictate protocol viability. A dominant restaking cartel decides which Actively Validated Services (AVSs) or consumer chains survive. New protocols must pay tribute to this cartel for security, replicating the VC gatekeeping of Web2.

Counter-movements are already forming. Protocols like Celestia and Avail champion sovereign rollups to escape this dynamic. Their data availability layers provide security without ceding consensus control, creating a political schism in modular design.

Evidence: The top 5 operators control over 60% of staked ETH. In a restaked future, this concentration grants them disproportionate power over hundreds of AVSs and L2s built on shared security layers.

takeaways
THE CENTRALIZATION PARADOX

Key Takeaways for Builders and Investors

Shared security models like restaking and shared sequencers promise scalability but create systemic power concentration.

01

The EigenLayer Cartel Problem

EigenLayer's $18B+ TVL creates a de facto security cartel. New L2s and AVSs must bid for this pooled capital, centralizing economic power.\n- Builder Risk: Your chain's security is now a commodity auctioned to the highest validator.\n- Investor Risk: Systemic failure of a major operator like Figment or Chorus One could cascade through hundreds of dependent chains.

$18B+
TVL
100+
AVSs
02

Shared Sequencer Monopolies (Espresso, Astria)

Outsourcing block production to a neutral third-party sequencer network creates a single point of control and failure.\n- Builder Lock-in: Your chain's liveness and censorship resistance depend on a ~5-10 entity committee.\n- Investor Blind Spot: Valuations ignore the long-term extractable value (LEV) captured by the sequencer layer, not the L2.

~500ms
Time to Finality
5-10
Key Entities
03

The Interoperability Bottleneck (LayerZero, Axelar)

Universal messaging layers become centralized choke points. Security is pooled, but control is not.\n- Builder Dependency: Your cross-chain app is only as secure as the ~31 Oracle/Optimistic Guardian nodes.\n- Investor Reality: These are high-margin tollbooth businesses, not decentralized protocols. $20M+ in annualized fees flow to a handful of entities.

31
Guardian Nodes
$20M+
Annual Fees
04

The Validator Centralization Flywheel

Shared security rewards scale, not diversity. Top-tier operators like Coinbase, Figment, and Kiln capture >60% of restaked ETH.\n- Builder Consequence: You inherit the political and regulatory risk of these centralized entities.\n- Investor Math: The staking yield is a mirage; real value accrues to the equity of the dominant operators.

>60%
Market Share
3-5
Dominant Firms
05

Solution: Purpose-Built Security with Economic Alignment

Avoid the commodity trap. Build chains where security is a core product feature, not a rented service.\n- For Builders: Use a dedicated validator set with token-economic skin in the game (e.g., dYdX, Sei). Accept lower initial TVL for sovereign security.\n- For Investors: Back teams building vertically integrated security stacks, not those outsourcing to EigenLayer for a quick launch.

0
Shared Risk
1:1
Economic Alignment
06

Solution: Modular but Sovereign Stacks (Celestia, EigenDA)

Decouple data availability and consensus from execution, but avoid bundling. Use competitive markets for each layer.\n- Builder Playbook: Use Celestia for DA, run your own sequencer, and choose a battle-tested bridge like Across or Connext.\n- Investor Thesis: Value accrual happens at the execution layer. Favor L2s that control their core infrastructure and can switch DA providers.

100x
Cheaper DA
Modular
Architecture
ENQUIRY

Get In Touch
today.

Our experts will offer a free quote and a 30min call to discuss your project.

NDA Protected
24h Response
Directly to Engineering Team
10+
Protocols Shipped
$20M+
TVL Overall
NDA Protected Directly to Engineering Team
Why Shared Security Models Will Centralize Power | ChainScore Blog