Censorship-resistance shifts to the application layer. In a rollup-centric world, L1 validators only see batched data, not individual transactions. The power to include, exclude, or reorder user ops now resides with the SUI (Sender User Intent) execution layer and its bundlers, like those in the Pimlico or Alchemy stacks.
account-abstraction-fixing-crypto-ux
Blog
The Future of Censorship-Resistance Lies with Bundlers
Account abstraction promises UX nirvana but centralizes power in bundlers. If they become OFAC-compliant choke points, they threaten the core value proposition of Ethereum. This is the next great crypto governance battle.
introduction
THE NEW GATEKEEPERS
Introduction
The censorship-resistance of user transactions is no longer determined by miners or validators, but by the economic actors who order them: the bundlers.
Bundlers are the new block producers. Unlike L1 miners, their profit motive is not pure block rewards but MEV extraction and fee arbitrage. This creates a perverse incentive to censor transactions that threaten their extractable value, a dynamic already visible in Flashbots auctions on Ethereum.
Account abstraction enables this shift. Standards like ERC-4337 formalize the bundler's role, making them a programmable, competitive marketplace. The censorship risk is not theoretical; protocols like UniswapX and CowSwap that rely on intents for routing are already dependent on these actors for fair execution.
key-trends
BUNDLERS AS THE NEW BATTLEGROUND
The Centralization Pressure Cooker
The censorship-resistance of user transactions is shifting from block builders to the entities that order them: bundlers. This creates a new centralization vector.
01
The Problem: Builder-Exclusive Order Flow
Today's PBS model funnels all transactions through a few dominant builders like Flashbots and BloXroute. This creates a single point of failure where censorship can be enforced at the protocol level, threatening the core promise of Ethereum.
- ~90%+ of blocks are built by the top 3 builders.
- Relayers can filter transactions based on OFAC lists.
- User intents are opaque and powerless at this layer.
90%+
Builder Dominance
1
Choke Point
02
The Solution: Bundlers as Censorship Shields
Bundlers in ERC-4337 and intent-centric architectures (like UniswapX and CowSwap) become the primary transaction orderers. A decentralized network of bundlers can route user operations around censoring builders, preserving access.
- Intent abstraction hides transaction details until execution.
- Competitive bundler market allows for redundant routing.
- Enables crypto-native MEV capture (e.g., via SUAVE) instead of off-chain auctions.
ERC-4337
Standard
Multi-Path
Routing
03
The New Risk: Bundler Cartels
If bundlers centralize (e.g., around a few wallet providers or infra giants like Alchemy or Blockdaemon), we simply recreate the builder problem at a higher layer. The economic design of bundling is critical to prevent this.
- Staking/PoS for bundlers (e.g., EigenLayer AVS) can align incentives.
- Permissionless inclusion lists must be enforceable by searchers.
- Requires credible neutrality in bundler software clients.
AVS
Staking Model
Cartel Risk
High
04
The Metric: Time-to-Censorship
The key measure of health is not just the number of bundlers, but the latency for a censored transaction to be included. A robust network should have a sub-5 minute time-to-inclusion for any valid user op, regardless of content.
- Requires active monitoring and fallback bundler circuits.
- Cross-chain intent systems (e.g., Across, LayerZero) add redundancy.
- This metric exposes the true cost of censorship for the network.
<5 min
Target TTC
Multi-Chain
Redundancy
DECENTRALIZATION FRONTIER
Bundler Market Share & Risk Profile
Comparative analysis of leading ERC-4337 bundler implementations, focusing on market dominance, technical architecture, and censorship-resistance guarantees.
| Metric / Feature | Pimlico (Paymaster) | Stackup | Alchemy | Ethereum Foundation (Reference) |
|---|---|---|---|---|
Current Market Share (UserOps) | ~45% | ~25% | ~15% | < 1% |
Execution Client Diversity | Geth, Nethermind | Geth, Erigon | Geth | Any |
Relay Network Nodes |
| ~500 permissioned | Centralized | N/A |
Max Extractable Value (MEV) Resistance | Flashbots Protect, SUAVE | Basic auction | None | None |
Censorship-Resistant Mode | ||||
Time to Finality (P95) | < 12 seconds | < 15 seconds | < 10 seconds |
|
Paymaster Integration | Native (Sponsorship) | Plugin-based | Limited API | None |
Avg. UserOp Fee Premium | 0.5% | 1.2% | 2.0% | Gas Cost Only |
deep-dive
THE CENTRALIZATION VECTOR
The Slippery Slope from Service to Censor
The architectural shift to rollups and account abstraction inadvertently concentrates censorship power in the hands of transaction ordering entities.
Bundlers control transaction ordering. Account Abstraction (ERC-4337) and rollup sequencers centralize the power to include, exclude, or reorder transactions. This creates a single point of failure for censorship, moving the attack surface from miners/validators to these new service providers.
The MEV supply chain is the pressure point. Proposer-Builder Separation (PBS) on Ethereum demonstrates that economic incentives corrupt neutrality. Bundlers like Stackup or Pimlico face the same extractive pressures as MEV builders, creating a financial motive to censor transactions for maximal extractable value.
Regulatory compliance becomes trivial to enforce. A sanctioned entity like Tornado Cash can be blacklisted at the bundler or sequencer level (e.g., Arbitrum, Optimism) with a simple software update. The service provider's legal survival will trump any ideological commitment to neutrality.
Evidence: Over 90% of Ethereum blocks are built by three MEV entities. This demonstrates the inevitable centralization of ordering power under financialized systems. Rollup sequencers operated by single entities are already the norm.
counter-argument
THE ARCHITECTURAL IMPERATIVE
The Counter-Argument: Permissionless Bundling Solves It
Censorship-resistance is not a validator problem; it is a market structure problem solved by competitive, permissionless bundling.
Permissionless bundling creates competition. Any actor can become a bundler, submitting transactions directly to the builder. This market dynamic forces inclusion. If one bundler censors, another economically incentivized bundler will include the transaction.
The builder is the bottleneck, not the bundler. The current censorship debate incorrectly focuses on validators. The real power resides with the block builder who assembles the payload. A permissionless bundler network is the counter-force to a censoring builder.
This is the Suave architecture. Projects like Flashbots' Suave explicitly decentralize the block building layer. Their model enables permissionless bundlers and searchers to operate, ensuring no single entity controls transaction flow.
Evidence: Ethereum's PBS trajectory. Proposer-Builder Separation (PBS) with permissionless bundling is the stated endgame. The ecosystem's development, from MEV-Boost to Suave, proves the technical path exists to harden censorship-resistance at the infrastructure layer.
protocol-spotlight
THE BUNDLER FRONTIER
Protocols Building the Anti-Censorship Stack
The censorship-resistance of Ethereum's base layer is compromised when centralized sequencers and RPC providers can filter transactions. The future lies with a new class of infrastructure: permissionless, competitive bundlers.
01
Flashbots SUAVE: The Decentralized Memory Pool
The Problem: Today's mempools are public and easily monitored, allowing MEV extraction and frontrunning.\nThe Solution: SUAVE creates a decentralized, encrypted mempool and a network of competitive builders.\n- Intent Expression: Users submit preferences (e.g., best price) instead of raw transactions.\n- MEV Redistribution: Competitive auction for block space returns value to users, not just searchers.
100%
Encrypted Flow
Multi-Chain
Scope
02
EigenLayer & AltLayer: Economic Security for Rollups
The Problem: New L2s and app-chains must bootstrap their own validator sets, often leading to centralized, censorable sequencers.\nThe Solution: Restaking pools like EigenLayer provide cryptoeconomic security for decentralized sequencer sets.\n- Shared Security: Borrow Ethereum's staked ETH to secure a rollup's sequencing.\n- Fast Finality: Networks like AltLayer leverage this for ~2s finality with decentralized fault proofs.
$15B+
Restaked TVL
~2s
Finality
03
The BloxRoute Gateway: Censorship-Resistant RPC
The Problem: RPC endpoints (Infura, Alchemy) can be forced to censor transactions via OFAC compliance, creating single points of failure.\nThe Solution: BloxRoute's Blockchain Distribution Network (BDN) is a peer-to-peer, geographically distributed network for transaction propagation.\n- Direct Peering: Bypasses centralized RPCs by connecting users directly to a global node network.\n- Guaranteed Inclusion: Uses proprietary protocols to ensure transactions reach miners/validators.
40%
Network Traffic
<100ms
Propagation
04
Anoma & Namada: The Intent-Centric Future
The Problem: Transaction-based models leak intent, creating MEV and requiring users to manage complex execution paths.\nThe Solution: A paradigm shift to intent-based architectures, where users declare what they want, not how to do it.\n- Privacy by Default: Shielded actions using zero-knowledge proofs.\n- Solver Competition: A network of solvers (like CowSwap, UniswapX) compete to fulfill intents optimally, improving price and censorship-resistance.
Intent
Paradigm
ZK-native
Architecture
05
Espresso Systems: Decentralizing the Sequencer
The Problem: Rollup sequencers (Arbitrum, Optimism) are currently centralized and can censor or reorder transactions.\nThe Solution: Espresso provides a shared, decentralized sequencing layer that rollups can opt into.\n- HotShot Consensus: A high-throughput PoS consensus protocol for fast block finality.\n- Interoperable MEV: Enables cross-rollup MEV opportunities and fair auction markets.
10k+
TPS Capacity
Shared
Sequencing
06
Shutter Network: Threshold-Encrypted Transactions
The Problem: Frontrunning bots scan the public mempool, extracting value from users' trades and governance votes.\nThe Solution: Uses threshold encryption (based on EigenLayer) to encrypt transactions until they are included in a block.\n- Blind Auctions: Enables fair, frontrunning-resistant DeFi auctions and governance.\n- Keyper Committee: A decentralized set of operators, slashed via EigenLayer for misbehavior, manages decryption keys.
Threshold
Encryption
Slashable
Keypers
future-outlook
THE BUNDLER
The Next Governance Battlefield
The future of censorship-resistance in rollups is shifting from sequencers to the permissionless bundler market.
Bundlers are the new gatekeepers. The sequencer's power to censor is neutralized by a competitive market of bundlers submitting transactions directly to the L1. This architectural shift in account abstraction (ERC-4337) moves the trust boundary.
Permissionless bundling defeats cartels. A single sequencer can filter transactions, but a decentralized network of bundlers like Pimlico, Stackup, and Alchemy creates redundancy. Censorship requires collusion across this entire economic layer.
The mempool is the battleground. Projects like EigenLayer and AltLayer are building shared, neutral mempools for intent settlement. This creates a public square for transactions that sequencers cannot ignore without sacrificing revenue.
Evidence: In a permissionless system, a user censored by one bundler has their transaction included by another within the same block. The economic cost of widespread censorship becomes infinite.
takeaways
THE BUNDLER FRONTIER
Key Takeaways for Builders and Investors
The MEV supply chain is the new battleground for censorship-resistance; builders must architect for it, and investors must back the infrastructure that wins.
01
The Problem: Validators Are the Weakest Link
Post-Merge, the validator set is the centralized choke point for transaction censorship. Relying on their altruism is a critical protocol failure.\n- Lido and Coinbase control >40% of Ethereum's stake.\n- OFAC-compliant blocks are a proven, ongoing threat to base-layer neutrality.
>40%
Stake Centralized
0
Guarantees
02
The Solution: Proposer-Builder Separation (PBS) is Non-Negotiable
PBS externalizes block building, creating a competitive market that separates profit motive from consensus. Censorship-resistance shifts to the builder/bundler layer.\n- Builders compete on inclusion, not compliance.\n- Enables encrypted mempools and threshold cryptography to blind validators.
1000+
Builder Pool
~1s
Auction Window
03
The Arbiter: SUAVE is the Endgame
Flashbots' SUAVE is building a decentralized, chain-agnostic block building network. It aims to be the universal preference layer for user transactions.\n- Decouples transaction ordering from any single chain's validators.\n- Creates a credibly neutral marketplace where the highest user preference wins, not just the highest bid.
Chain-Agnostic
Scope
TBD
Dominance
04
The Tactic: Encrypted Mempools via Threshold Cryptography
To prevent frontrunning and censorship, transaction content must be hidden from builders and validators until inclusion. This is a cryptographic, not social, solution.\n- Shamir's Secret Sharing splits transactions across a committee.\n- Enables fair ordering and neutral fee markets, critical for the next billion users.
~500ms
Decryption Latency
N of K
Trust Model
05
The Market: Intent-Based Architectures Will Dominate
Users will declare outcomes, not sign raw transactions. Systems like UniswapX, CowSwap, and Across abstract complexity into solver networks.\n- Bundlers become solvers, competing on fulfillment quality and cost.\n- Creates a $10B+ market for execution quality, separate from L1 gas fees.
$10B+
Market Size
90%+
Efficiency Gain
06
The Bet: Invest in the MEV Supply Chain
The value accrual is shifting from L1 tokens to the infrastructure that orders, bundles, and executes. This is where the next infrastructure unicorns are being built.\n- Back teams building generalized PBS, secure enclaves, and solver networks.\n- The moat is in cross-chain liquidity integration and cryptographic innovation.
10x
Value Shift
Infrastructure
Accrual Layer
ENQUIRY
Get In Touch
today.
Our experts will offer a free quote and a 30min call to discuss your project.
NDA Protected
Confidentiality24h Response
Time to ValueDirectly to Engineering Team
No Sales Fluff10+
Protocols Shipped
$20M+
TVL Overall