Ethereum Research Post Exposes Sybil Flaw in AUCIL Design
A new ethereum research post published on July 12, 2026 identifies specific ways that sybil attacks can undermine AUCIL, the auction-based inclusion list design proposed as an alternative to the FOCIL mechanism heading into Ethereum’s Hegota upgrade. The post by researcher Abhimanyu Nag, accompanied by a full technical analysis on functor.network, demonstrates that an attacker controlling as few as two proposer identities can manipulate AUCIL’s fee allocation in ways that reduce censorship resistance guarantees. The finding fills a research gap that the author explicitly flags: while bribery attacks on FOCIL received a comprehensive treatment in a 2025 paper by Stouka, Ma, and Thiery, no corresponding sybil analysis existed for AUCIL until now.
The timing is significant. FOCIL, formally EIP-7805, was confirmed as the consensus-layer headliner for the Hegota hard fork during an All Core Devs call earlier this year, with deployment targeted for the second half of 2026. AUCIL remains at the research stage. But because both designs attempt to solve the same problem, and because several of the same researchers contributed to both, understanding where each breaks is directly relevant to Ethereum’s roadmap.
Why Ethereum Needs Inclusion Lists
The core problem is block builder concentration. Under Ethereum’s current proposer-builder separation (PBS) architecture, validators outsource block construction to specialized builders who compete through MEV-Boost relays. Research from early 2026 found that just five builders control roughly 97% of all MEV-Boost blocks, with the top three accounting for approximately 80% of production. That concentration gives a small number of entities significant power over which transactions get included and which get delayed or excluded.
After the US Treasury sanctioned Tornado Cash in 2022, some major relays began filtering transactions associated with sanctioned addresses. At peak, OFAC-compliant relays handled nearly 80% of Ethereum blocks. That figure has since declined, but the structural risk remains: a handful of builders can effectively censor valid transactions at the block production level.
Inclusion lists are Ethereum’s proposed solution. Rather than trusting builders to include all valid transactions, the protocol would give a separate group of participants the power to enforce inclusion. Two major designs have emerged.
FOCIL vs AUCIL
FOCIL (Fork-Choice Enforced Inclusion Lists) takes a consensus-based approach. Each slot, a committee of 16 validators is chosen at random. These validators independently scan the public mempool and each assemble a short list of pending transactions, with each list capped at 8 kilobytes. The committee broadcasts these lists, and the proposer of the next block must include the specified transactions. If a block ignores valid inclusion list transactions, attesters will not vote for it, effectively rejecting it from the canonical chain.
The key property: because committee members are selected randomly, a sybil attacker cannot increase their representation by creating fake identities. The committee selection mechanism is inherently sybil-resistant.
AUCIL, developed by Sarisht Wadhwa, Julian Ma, Thomas Thiery, Barnabé Monnot, and collaborators across Duke University, Yale, and the Ethereum Foundation, takes a different path. Rather than relying on random committee selection, AUCIL structures participation around economic competition. Proposers submit input lists, and a greedy algorithm computes a recommended allocation that maximizes expected utility. Aggregation into a final inclusion list happens through a scoring auction, where candidates compete based on how many available input lists they aggregate.
The original AUCIL paper, published in February 2025, proved that censorship through the aggregation stage costs at least a quantifiable minimum amount, providing a formal lower bound on bribery costs.
| Feature | FOCIL (EIP-7805) | AUCIL |
| Mechanism | Committee-based, fork-choice enforced | Auction-based, economic incentives |
| Committee selection | Random validators each slot | Strategic participation with fee-driven allocation |
| Sybil resistance | Inherent at committee selection | Under analysis, vectors identified |
| Fee mechanism | No explicit fees | Auction pricing split among contributing proposers |
| Status | Scheduled for Hegota, H2 2026 | Research stage |
| Key researchers | Thomas Thiery, Francesco D’Amato, Julian Ma | Sarisht Wadhwa, Kartik Nayak, Barnabé Monnot |
Where Sybil Attacks Enter AUCIL
The new analysis identifies the attack surface in AUCIL’s first phase, the input list allocation.
In AUCIL, transaction fees are split among the proposers assigned to the same transaction. The greedy allocation algorithm assigns transactions based on marginal payoff, and the resulting equilibrium is designed to be individually incentive-compatible (IC), meaning no single proposer benefits from unilaterally deviating.
Nag’s analysis shows that individual IC does not imply sybil-proofness. The functor.network post presents a concrete numerical example. An attacker controlling two proposer identities can coordinate their deviation: one identity switches its assigned transaction, accepting a personal loss, while the other identity’s reward increases by a larger amount. The net effect is positive for the attacker.
More critically, the attack reduces the multiplicity of a transaction, meaning the number of input lists that include it. In the paper’s example, one transaction’s multiplicity drops from two to one. Since AUCIL’s censorship resistance guarantee depends on transactions appearing across multiple input lists, reducing multiplicity directly weakens the protocol’s ability to prevent censorship.
The analysis also recalculates AUCIL’s original bribery cost guarantee under sybil conditions. The lower bound on censorship cost, which the original paper established as a formal security property, no longer holds when the attacker controls multiple identities. The exact new cost depends on what replacement transactions are available to each proposer.
What the Findings Suggest About Fixing AUCIL
The post does not argue that AUCIL is fatally flawed. The author explicitly frames the analysis as early-stage, noting that a full theory of sybil-proof auction-based inclusion lists remains unwritten.
Two directions are proposed. The first is replacing AUCIL’s current fee-sharing mechanism with stake-based partitions, where fees are distributed proportionally to each proposer’s staked ETH rather than split equally among identities. Because sybil identities cannot multiply real stake, this would remove the incentive to split into multiple fake participants. The second suggestion involves redesigning the scoring auction in AUCIL’s aggregation phase, potentially through a stake-weighted lottery that replaces the current VRF-based random bias.
Both suggestions require formal analysis before they could become implementable proposals.
The Broader Ethereum Research Landscape
This analysis arrives during a period of structural change in how Ethereum research is organized. Five former senior Ethereum Foundation researchers, including Barnabé Monnot, launched Ethlabs on June 22 as an independent nonprofit R&D lab backed by Bitmine, SharpLink, and Ethereum co-founder Joe Lubin. Monnot is a co-author of the original AUCIL paper, and his departure is part of a broader wave of at least ten senior exits from the Ethereum Foundation in 2026.
That Monnot helped design AUCIL while at the Ethereum Foundation and now leads research at an independent organization illustrates a point the ethresear.ch post implicitly makes: this kind of critical analysis continues regardless of organizational boundaries. Abhimanyu Nag’s post credits initial discussions with Sarisht Wadhwa and the AUCIL collaborators, suggesting the research community itself welcomes scrutiny of its own designs.
Meanwhile, FOCIL implementation is progressing. Client teams have begun work across Ethereum’s consensus and execution layers, with Ethereum researcher Thomas Thiery tracking milestones through a public dashboard. Vitalik Buterin has described FOCIL as enabling censorship-resistant inclusion of any valid transaction within one to two slots, and the Lean Ethereum roadmap published July 4 positions the Hegota upgrade as one step in a broader multi-year overhaul of the network’s core infrastructure.
What to Watch
The ethresear.ch post raises questions but does not settle them. Whether the sybil vulnerability in AUCIL prompts a formal revision, a new design variant, or simply reinforces the community’s confidence in FOCIL as the near-term choice will depend on follow-up research.
The variables worth tracking are direct: whether AUCIL’s authors publish a response addressing the sybil vectors, whether stake-based fee partitions receive formal modeling, and whether Hegota’s FOCIL implementation encounters its own unexpected edge cases during testnet deployment. The sybil analysis itself is a normal part of how open protocol research advances. Identifying attack surfaces before deployment is the intended function of forums like ethresear.ch, and the fact that AUCIL’s sybil properties are being examined while FOCIL heads toward production suggests the research process is working as designed.
FAQ
What is a sybil attack in blockchain?
A sybil attack occurs when a single entity creates multiple fake identities to gain disproportionate influence over a decentralized system. In the context of inclusion lists, a sybil attacker controlling multiple proposer identities can manipulate fee allocation and weaken censorship resistance guarantees without spending additional resources.
What is FOCIL on Ethereum?
FOCIL stands for Fork-Choice Enforced Inclusion Lists, formally designated as EIP-7805. It randomly selects 16 validators per slot to create lists of transactions that must be included in the next block. It is scheduled as the consensus-layer headliner for Ethereum’s Hegota upgrade in the second half of 2026.
What is the AUCIL inclusion list design?
AUCIL is an auction-based inclusion list mechanism developed by researchers at Duke University, Yale, and the Ethereum Foundation. It uses economic incentives rather than random selection, with proposers competing through a scoring auction to aggregate transaction lists. AUCIL remains at the research stage and is not scheduled for any upgrade.
When is the Ethereum Hegota upgrade?
The Hegota hard fork is targeted for the second half of 2026, following the Glamsterdam upgrade. As of mid-July 2026, no final mainnet activation date has been confirmed. EIP-7805 (FOCIL) is the only proposal officially marked as scheduled for inclusion.
