Rational Censorship Attack: Breaking Blockchain with a Blackboard
Michelle Yeo, Haoqian Zhang · 2025
Read Rational Censorship Attack: Breaking Blockchain with a Blackboard by Michelle Yeo, Haoqian Zhang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-01453.txt.
Introduction
Michelle Yeo and Haoqian Zhang presented a rational censorship attack on blockchains in 2025. Blockchains rest on the idea that no group can easily block others' transactions. The authors view chain security through game theory, where every user acts in their own interest.
In the attack, a group with enough voting power censors the rest and takes all chain rewards for itself. Coordination needs only a public board that anyone can read and write. The authors model the attack as a formal game. They prove that joining the attack, and honestly stating one's voting power, forms a stable equilibrium of that game. Countermeasures and lessons for users and designers are also discussed.
Catalog record
| CREATOR | Michelle Yeo, Haoqian Zhang |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 62 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-01453
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Rational Censorship Attack: Breaking Blockchain with a Blackboard Michelle Yeo, Haoqian Zhang Abstract Censorship resilience is a fundamental assumption underlying the security of blockchain protocols. Additionally, the analysis of blockchain security from an economic and game theoretic perspective has been growing in popularity in recent years. In this work, we present a surprising rational censorship attack on blockchain censorship resilience when we adopt the analysis of blockchain security from a game theoretic lens and assume all users are rational. In our attack, a colluding group with sufficient voting power censors the remainder nodes such that the group alone can gain all the rewards from maintaining the blockchain. We show that if nodes are rational, coordinating this attack just requires a public read and write blackboard and we formally model the attack using a game theoretic framework. Furthermore, we note that to ensure the success of the attack, nodes need to know the total true voting power held by the colluding group. We prove that the strategy to join the rational censorship attack and also for nodes to honestly declare their power is a subgame perfect equilibrium in the corresponding extensive form game induced by our attack. Finally, we discuss the implications of the attack on blockchain users and protocol designers as well as some potential countermeasures. 1 Introduction Blockchains like Bitcoin and Ethereum allow users to transact with others in a decentralized fashion without going through a central entity like a bank. A key security goal of such decentralized systems is censorship resilience: any user should in theory be able to participate in both consensus as well as the contribution of financial transactions to the blockchain. Although transaction-level censorship has been well studied with several defensive measures proposed [ 5 , 6 , 13 , 17 , 19 , 33 , 35 , 41 ] , the analysis of consensus-level censorship attacks and corresponding mitigation is a lot more sparse [ 16 , 38 , 40 ] . Indeed, two commonly proposed reasons why mitigating consensus-level censorship is not such a pressing issue is, firstly, that the amount of cost involved to coordinate such an attack is typically substantial. Indeed, an attacker not only has to gather sufficient power to launch the attack, but also has to find ways to coordinate strategies with other attacking nodes [ 2 ] . Secondly, it has also been argued that the short-term monetary benefits of attacking consensus on the blockchain could be outweighed by the long-term consequences of the attack, which includes a potential drop in the value of the underlying cryptocurrency resulting from the reduced trust in the blockchain [ 12 , 25 ] . On an orthogonal direction of research, recent years have seen a shift in the analysis of blockchain security from a cryptographic or worst-case perspective to an economic or game theoretic perspective (see e.g., [ 11 , 23 , 36 , 45 , 46 ] and more). In contrast to the classic cryptographic analysis where users are classified into the binary categories of byzantine or honest, an economic or game theoretic analyses typically assume all users in the blockchain are rational, that is, they are profit-maximizing. Under this perspective, a given system with rational users is shown to insecure if there exists attacks such that the revenue of the attack outweighs the cost of launching the attack. The shift from cryptographic to rational analysis in blockchains is mainly motivated by the argument that in real world systems users might be motivated by economic reasons instead of purely byzantine ones to deviate from honest behavior. Hence, not only does this harmonizes well with blockchain security analysis as blockchains typically contain financial data, but rational analysis also allows for a more subtle perspective on blockchain security, where an argument for the security of the blockchain could still be made under rational arguments when the blockchain would be deemed insecure under cryptographic ones. Interestingly, adopting the rational perspective of blockchain security could lead to arguments weakening the censorship resilience of blockchains. For instance, in [ 25 ] the authors point out that assuming blockchains exist in a larger financial ecosystem, there could be rational reasons to attack some blockchain system if one short sells the underlying cryptocurrency. Thus, although the value of the cryptocurrency could diminish from attacks against the blockchain, a user that short sells sufficient amount of the currency would still make an overall profit in the larger financial ecosystem. This weakens the aforementioned second argument for consensus-level blockchain censorship resistance, i.e., that the short term monetary benefits of censorship attacks are overwhelmed by their long-term financial impact. Motivated by this observation, we focus on the first argument for consensus-level censorship resistance, i.e., the substantial cost involved in launching censorship attacks, and show that adopting the rational perspective makes it extremely cheap to launch a majority attack on the blockchain, which we term the rational censorship attack . The rational censorship attack involves users colluding together and forming a coalition that controls sufficient voting power. Then, the colluding group simply ignores messages from the remaining excluded users, effectively removing them from participating in blockchain consensus and hence censoring them. The surprising novelty of our attack lies in its simplicity. Indeed, coordinating our attack does not involve on any form of private communication between users in the blockchain except for public messages posted on a public blackboard with read and write access. In fact, we show that a single message posted on the blackboard is sufficient to initiate the attack. We also show that the knowledge that all other users will attempt to join the attacking coalition is sufficient incentive for any user to also join the coalition and attack the blockchain. 1.1 Contributions Our contributions are summarized as follows: • We present a novel simple rational censorship attack on blockchains that guarantees a larger payoff for attacking nodes with communication only via a public read-and-write blackboard. • We formally model our attack using a game theoretic framework and prove that attacking the blockchain is a subgame perfect equilibrium in the extensive form game induced by our attack ( Theorem 2.2 in Section 2.4 ). • We further prove that our attack strategy induces truthfulness among rational nodes, that is, nodes are incentivized to declare their true voting power for the attack. • We present two plausible countermeasures to the attack in Section 3 and discuss their effectiveness. • We discuss the implications of the attack on blockchain users and protocol designers in Section 4 . 1.2 Related work Attacks on blockchain security leveraging rationality. The earliest work that exposes vulnerabilities in blockchain security by leveraging rationality arguments is selfish mining by Eyal and Sirer, demonstrating that the Bitcoin protocol is not incentive-compatible [ 23 ] . Specifically, the work shows that even if the adversary controls less than 50% of the hashing power in Bitcoin, they can still launch a selfish mining attack to reap a larger fraction of rewards compared to the honest strategy of following the stipulated protocol. Following the initial selfish mining attack analysis, several follow-up works present other attacks that expose further vulnerabilities behind the incentive mechanisms grounding blockchain protocols, for instance whale attacks [