Ethereum pulls post-quantum target to December 2028 at Devcon 8

Ethereum Foundation researcher Justin Drake told Devcon 8 in Mumbai on November 3 that the foundation is pulling its post-quantum transition target to December 2028, one year ahead of the December 2029 date the September 2025 roadmap set. "Our hope is that by December 2028 we'll be able to wrap up the full post-quantum and post-AI transition," Drake said, per Crypto Briefing's conference writeup.

The plan is to ship three network upgrades within 24 months of an early-2027 milestone that the foundation has not yet named publicly.

What has to be replaced

Two signature schemes in production today rely on elliptic-curve assumptions that a sufficiently large fault-tolerant quantum computer would break in polynomial time:

• ECDSA (secp256k1) — secures user accounts. About 72% of ETH supply sits in accounts whose public keys are already exposed on-chain, per Crypto Briefing's figure from the Devcon talk; a reveal-on-spend address loses that protection once it has sent a transaction.
• BLS12-381 — used by validators for consensus signatures and aggregation.

The foundation's Post-Quantum Security team, formed January 2026, is standardizing on hash-based replacements: leanSPHINCS and leanXMSS. Both rely on the security of the underlying hash function rather than any number-theoretic hardness assumption, which is why they survive Shor's algorithm. The lean prefixes refer to signature-size and verification-cost optimizations the EF team has been working toward on-chain practicality for; a standalone SPHINCS+ signature runs to tens of kilobytes, which is a lot to put in every transaction.

The direction is consistent with the foundation's August 2026 roadmap note, which also dropped Poseidon as the SNARK-friendly hash target in favor of standardized hashes like SHA and BLAKE, and with the broader post-quantum signature work tracked by Everstake.

Why pull the date in

Drake flagged in October that progress in generative-model-assisted cryptanalysis is likely to compress the usable timeline for elliptic-curve schemes; the foundation's working assumption is that the post-quantum transition has to be done before a cryptographically relevant quantum machine ships, not immediately after. Vitalik Buterin warned in November 2025 that elliptic cryptography could become vulnerable by 2028 — the new target now lines up with that warning rather than the 2029 outer bound.

What ships when

• Early 2027: an unnamed milestone. The foundation has not published the EIP set.
• 2027–2028: three network upgrades on the 24-month clock. Scope covers consensus (BLS → post-quantum aggregation), execution (ECDSA alternatives for user accounts, likely opt-in first), and data availability.
• December 2028: full transition target.

No specific EIP numbers have been published for the leanSPHINCS/leanXMSS proposals yet. The opt-in account-level variant SPHINCS- has been discussed in forum threads but not formally proposed as an EIP.

What to watch

1. The early-2027 milestone's EIP numbers — once those land, the three-upgrade sequence becomes a dated list rather than a target.
2. Whether exchanges and custodians commit to a key-migration timeline. About 72% of ETH is in exposed-key accounts today; moving that balance would take months of user-driven signing operations.
3. Rollup teams' positions. Many L2s inherit the base-layer signature scheme by reference; a post-quantum L1 does not automatically give post-quantum L2s.
4. BLS aggregation performance for leanSPHINCS. The consensus layer currently aggregates hundreds of thousands of signatures per slot; a hash-based scheme without a workable aggregation construction would add bandwidth pressure to the attestation subnets.

Context

Ethereum is not alone. Bitcoin Core developers have been moving toward opt-in post-quantum output types, and Starknet has backed an independent post-quantum L1 rather than wait on the base layer it settles to. Devcon 8 fixes Ethereum's position in that field: a hash-based signature stack, lean enough to live on-chain, shipped in three upgrades over 24 months, with a 2028 deadline the foundation is now treating as a public commitment rather than an aspirational one.