Ethereum co-founder Vitalik Buterin warned on October 7, 2026 that AI-accelerated mathematical research has a "good chance" of significantly weakening lattice-based cryptography within two years. The warning covers ML-DSA, the lattice-based signature scheme, and fully homomorphic encryption (FHE), and extends to ECDSA, the elliptic curve signature algorithm used across most of the industry today.

Buterin's concern targets a family of cryptography that the industry has largely treated as its quantum-era safe harbor. Lattice schemes rely on hard geometric problems involving grids of points in many dimensions. FHE lets computers process encrypted data without decrypting it, a property that underpins a range of privacy applications. If AI-driven math research erodes these constructions before quantum computers even arrive, protocols could face migration pressure away from the very tools positioned as post-quantum defenses.

What did Buterin recommend?

He paired the warning with a set of concrete operational recommendations:

• Conservative lattice parameters. Where lattice schemes remain in use, he suggested conservative settings, including a 10-fold increase in key sizes where applicable.
• No encrypted data stored directly on-chain. Data posted to a blockchain persists permanently; if the encryption protecting it weakens later, the privacy is lost for good.
• Key rotation for multisigs. Multisig signers should switch to new keys after each operation.
• Offchain signature gathering. Multisigs should collect signatures offchain where possible to reduce exposure.
• Caution on mass fund migrations. He discouraged large-scale moves of funds, which carry their own risks.

The multisig guidance connects to a broader point about public keys. Exposing a public key on-chain gives future attackers a fixed target to work against. As AI and cryptanalysis techniques advance, that exposure becomes a growing liability.

Why hash-based signatures?

Buterin described a move away from dependence on lattice cryptography toward pure hash-based signature schemes, naming WOTS and SPHINCS+. Hash-based signatures rest on a simpler foundation, with security relying on hash functions rather than exotic mathematical assumptions. Fewer assumptions means fewer places for a clever new attack to land.

WOTS, the Winternitz one-time signature, is designed so each key signs only once. That design philosophy aligns directly with his advice that multisig signers rotate keys after each operation.

The warning also inverts a position Buterin took in September 2026, when he projected that AI-assisted formal verification could benefit cybersecurity. Formal verification uses mathematical proofs to confirm that code behaves as intended. The same capability that helps prove code correct may also help prove that certain encryption is weaker than advertised. The technology cuts both ways.

What are the operational consequences?

No protocol exploits or acute incidents followed the warning; the discussion has centered on long-term cryptographic strategy rather than any immediate event. The implications split across three constituencies.

For builders, teams relying on lattice-based tools — including FHE for privacy features — face a reason to revisit parameter choices. Projects tempted to store encrypted data on-chain now have a prominent voice arguing against the practice outright.

For multisig operators, including treasuries, DAOs and custody setups, the advice is operational: rotate keys after each use, gather signatures offchain, and minimize how often public keys hit the chain.

For protocol architects, the larger question is migration sequencing. Buterin specifically cautioned that large-scale fund migrations carry their own risks, and a rushed transfer can create the very exposure it aims to avoid. He holds the majority of his net worth in crypto assets, which places his own holdings on the line if signature schemes falter — a fact that colors his caution about hasty relocations.

The competitive variable now is timing. Ethereum's transition toward hash-based signatures such as WOTS and SPHINCS+ is a long-term roadmap project, while Buterin's window for serious lattice weakening is two years. Whether the migration moves faster than the AI math research it races against will determine how much of today's lattice-based infrastructure survives intact.