Cardano创始人霍金森挑战以太坊创始人维塔利克：AI对加密货币的威胁

卡达诺创始人霍金森反驳以太坊联合创始人巴特林的AI安全警告

卡达诺（Cardano）创始人查尔斯·霍金森（Charles Hoskinson）对以太坊联合创始人维塔利克·布特林（Vitalik Buterin）提出的警告进行了反驳。布特林曾警告称，人工智能的进步可能会削弱抗量子密码系统的安全性，而霍金森则辩称，现有的安全标准已经考虑了数十年的数学攻击风险。

关于格密码与AI威胁的争议

此次争论始于布特林发出警告，指出基于人工智能辅助的数学研究可能挑战当前被认为安全的密码系统的假设。他特别指出了基于格的密码学（lattice-based cryptography），认为日益强大的AI系统可能会发现解决用于保护数字信息的数学问题的更高效方法。在他看来，自动化数学研究的进步可以将几十年的研究成果压缩到更短的时间内，从而迫使开发人员重新评估现有算法的安全性。

布特林将这种潜在的发展比作整数分解领域的进展，研究人员发现了使某些计算比早期估计更容易的方法。他认为基于格的安全性可能会经历类似的突破，导致当前的参数选择不足。

然而，霍金森拒绝了这一类比。他指出，整数分解的历史并不能证明格系统存在同等弱点。“因子分解的类比是错误的，”霍金森写道，解释说整数分解的改进依赖于特定的数学技术，而这些技术在现代密码学使用的格问题上尚未得到证明。卡达诺创始人回顾了自1982年Lenstra–Lenstra–Lovász（LLL）格约减算法引入以来的研究，并引用随后的格筛法和其他攻击方法的改进作为数十年测试的例子。据霍金森称，现代基于格的算法所使用的安全参数已经考虑了这些技术的已知进步。

捍卫ML-KEM与ML-DSA标准

霍金森论点的核心涉及ML-KEM和ML-DSA，这两个由美国国家标准与技术研究院（NIST）标准化的密码系统。ML-KEM允许各方建立共享的加密密钥，而ML-DSA提供用于验证信息和授权的身份验证数字签名。NIST于2024年8月在多年的研究和公开评估后批准了这两种系统（分别符合FIPS 203和FIPS 204标准）。此外，NIST还单独标准化了基于哈希的数字签名系统SLH-DSA（符合FIPS 205标准）。

Hodgson argued that ML-KEM and ML-DSA were selected after researchers reviewed different methods of attacking lattice-based mathematical problems. He maintains that the existence of mathematical structures does not automatically mean that the algorithm is insecure, especially when the security parameters are selected using established attack cost calculations. Previously, Butrin had suggested that future AI advances might require a significant increase in the size of grid-based keys, even by ten times. Hodgson rejected the proposal, calling it "numerology" because the proposed increase did not have specific algorithms or calculations to support it. According to Cadano's founders, choosing a larger key without identifying the corresponding attack does not provide a reliable way to measure security. He believes researchers should first determine the computational cost of new attacks before deciding whether existing encryption parameters need to be changed.

NIST currently describes ML-KEM and ML-DSA as algorithms that can withstand sufficiently powerful quantum computer attacks. These standards do not establish that these systems are immune to all possible mathematical discoveries.

Ethereum's hash-based security plan faces doubts

Butrin's concerns are related to Ethereum's work to design cryptographic systems that are resistant to future quantum computing attacks. Ethereum's "Lean Ethereum" research is increasingly leaning towards the use of hash-based signatures, including WOTS and SPHINCS+ constructs, for application in scenarios where developers can use them. Butrin believes that hash-based methods may provide greater long-term security because they do not rely on the same mathematical assumptions as lattice-based systems.

然而，霍金森挑战了“哈希函数因具有较少可利用的结构而自动更安全”的观点。他指出MD5和SHA-1这两个哈希算法的历史弱点，研究人员通过其内部设计的缺陷成功攻击了它们。霍金森表示，即使SHA-256也没有数学证明确立其对所有可能攻击的免疫性。“它的安全性在于至今无人破解它，”他写道，主张应对各类密码算法适用相同的审查标准。他还质疑以太坊之前对Poseidon和Poseidon2的研究，这两种哈希函数旨在高效用于某些区块链证明系统中。以太坊研究人员已经在他们的Layer-1路线图中弃用了Poseidon，转而 favor 诸如SHA-2和BLAKE2s等成熟替代方案。今年8月，以太坊基金会研究员贾斯汀·德雷克（Justin Drake）解释称，二元域证明系统的改进使得传统哈希在未来的网络设计中更具实用性。

霍金森认为，在哈希函数的局限性可接受的场景中，基于哈希的签名仍然有用，但它们不能取代公钥密码学执行的所有功能。这场分歧还延伸到了用于安全消息传递、网站及其他需要在各方之间建立受保护通信的应用程序的加密系统。

AI密码学辩论波及区块链钱包安全

The confrontation between the two founders comes as blockchain developers consider how to protect digital assets from future advances in AI and quantum computing. Butrin recommends that users consider storing cryptocurrencies at addresses where public keys are not exposed through outgoing transactions, provided this can be done securely. He emphasized that the recommendation did not justify the hasty relocation of wallets, as it itself could pose security risks.

Ethereum founders support plans to make Ethereum resistant to quantum attacks, including changes to transaction signatures, consensus mechanisms and zero-knowledge proof systems. Hodgson insisted that research on lattice-based methods should be carried out in parallel with research on hash-based alternatives.

The debate has affected companies that develop institutional hosting systems. In May, BitGo tested anti-quantum wallet signatures using ML-DSA technology co-developed with Silence Laboratories. The demonstration examined whether post-quantum signatures could work in existing institutional wallet programs. This is a controlled test and is not evidence that the current wallet password has been cracked. Other blockchain networks are also working hard to provide similar protection. Sui's anti-quantum security roadmap includes lattice-based and hash-based signature options for future account authentication.

Hodgson did not announce a new Cadano encryption upgrade or deployment timetable in his October 9 response. He called for any newly discovered attacks to be tested, cost analyzed and reviewed before developers modify security parameters or abandon established algorithms.