Charles Hoskinson, founder of Cardano, has contended that Ethereum co-founder Vitalik Buterin is harming quantum-resistant cryptography through speculative cautions concerning artificial intelligence-driven mathematical breakthroughs.
Responding in an Oct. 9 post, Hoskinson pushed back against Buterin’s doubts regarding lattice-based cryptography. He maintained that decades of safety research have already accounted for the known vulnerabilities of that technology.
He cautioned that prompting developers to steer clear of established methods might delay the integration of safeguards currently being rolled out across internet infrastructure, which could leave networks vulnerable to upcoming quantum threats.
This backlash comes after Buterin cautioned that AI-aided mathematical breakthroughs might undermine encryption built to withstand quantum computing.
Buterin stated that Ethereum’s extended “lean” strategy has shifted its focus to hash-based signatures and proofs while bypassing lattice-based architectures.
Dismissing that logic, Hoskinson claimed Buterin is overly committed to Ethereum’s current research path to re-evaluate the strategy.
“The case against lattices is the GNFS story, a.k.a. a hunch about ‘structure,’ and a multiplier pulled out of thin air,” Hoskinson stated.
At the center of the disagreement are post-quantum substitutes for the elliptic-curve signatures currently relied upon by Bitcoin and Ethereum.
Hoskinson challenges the mathematical case against lattices
Buterin’s argument leans in part on the history of integer factorization, where mathematical leaps—such as the general number field sieve—vastly enhanced methods used to target RSA encryption.
He proposed that artificial intelligence might compress decades of equivalent mathematical evolution into a much tighter timeframe, potentially uncovering unforeseen shortcuts to compromise lattice-dependent frameworks.
Hoskinson countered that the number field sieve arose from methodologies tied to arithmetic relationships and smooth numbers, noting that no parallel mechanism has been proven to threaten the lattice problems supporting current post-quantum frameworks.
The Cardano creator highlighted over 40 years of study, featuring progress in lattice reduction and sieving algorithms, which have steadily refined attacks without yielding a broad breakthrough capable of breaking properly secured systems.
In 2024, the US National Institute of Standards and Technology standardized ML-KEM for key encapsulation and ML-DSA for digital signatures. Hoskinson pointed out that their security settings factor in these recognized attacks.
Furthermore, Hoskinson dismissed Buterin’s hypothetical proposal that scaling key sizes by a factor of ten could safeguard public-key encryption from AI-powered mathematical advances.
Describing the idea to “Multiply key sizes by ten” as “numerology,” Hoskinson argued that defense parameters must instead scale based on quantifiable enhancements in attack algorithms.
While an increase in attack efficiency might call for slightly expanded parameters, a core mathematical breakthrough would necessitate swapping out an algorithm entirely.
Ethereum’s hash-based strategy faces its own questions
Hoskinson likewise questioned the premise that hash-based cryptography delivers superior protection against unexpected mathematical revelations.
He referenced past vulnerabilities found in MD5 and SHA-1 to show that hash functions can harbor exploitable designs, though those past flaws do not point to weaknesses within contemporary designs like SHA-256.
His critique further touched on Poseidon and Poseidon2, which are hash functions optimized for zero-knowledge proofs and tied to Ethereum’s broader cryptographic studies.
The Ethereum Foundation has backed studies examining Poseidon’s resilience against algebraic attacks, which include analyses utilizing Gröbner bases alongside other cryptanalytic strategies.
Hoskinson asserted that these frameworks are similarly vulnerable targets for AI-assisted mathematical findings, questioning why lattice systems are singled out for heightened skepticism.
While hash-based signatures supply quantum-resistant transaction verification, lattice methodologies additionally back key encapsulation for encryption along with other advanced cryptographic mechanisms.
Consequently, walking away from lattice research might restrict the toolset available to engineers designing privacy platforms, secure messaging, and alternative systems that rely on those functions.