Ethereum researchers have put forward an early draft that would rebuild the smart contract used to admit new validators, creating a path to quantum-resistant staking keys before today’s cryptography becomes a liability.
The plan focuses on the deposit contract, the gateway through which operators lock ether and join consensus.
About 42.4 million ETH—worth roughly $104 billion at recent prices—is already staked, and every one of those positions is guarded by Boneh–Lynn–Shacham (BLS) keys. BLS is efficient because many validator signatures can be compressed into a single proof.
It also rests on elliptic-curve math that a sufficiently capable quantum machine running Shor’s algorithm could, in principle, invert, allowing forged attestations.
The existing contract hard-codes BLS dimensions: 48-byte public keys and 96-byte signatures.
Even if the community agreed on a post-quantum scheme tomorrow, the contract could not accept it.
The draft, filed as pull request 12235 and informally labeled EIP-8394, would replace that rigid layout with variable-length credentials of up to 8,192 bytes and a scheme tag on every deposit.
Tag zero remains BLS; later tags are reserved for whatever standard developers later adopt.
The contract would start in a mode that still accepts BLS deposits, then allow a one-way protocol switch that permanently closes new BLS onboarding.
Existing validators would keep operating; only new joiners would be forced onto the next format.
That switch is infrastructure, not a finished cryptographic swap.
A separate consensus-layer upgrade would still have to define how clients verify the new signatures.
Researchers have discussed hash-based designs such as leanXMSS, often paired with a small zero-knowledge virtual machine to compress bulky proofs so the network stays cheap to run.
Account-level agility is moving on a parallel track through EIP-8141, which would let ordinary wallets change their signing scheme without changing addresses.
The Ethereum Foundation has treated post-quantum work as a priority since early 2026 and has publicly aimed at around 2029 for the core protocol pieces.
The timing is deliberate.
Cryptographic migrations on a live, multi-client chain take years: client teams, staking pools, liquid-staking protocols, and restaking platforms all have to move in lockstep.
Google Quantum AI’s 2026 analysis sketched several attack paths against Ethereum-related assets and helped pull the threat out of the “someday” category.
Estimates of when elliptic-curve schemes might fall still span the late 2020s to mid-2030s, but protocol designers cannot wait for certainty. Starting late would mean finishing after the window has already closed.
The draft is not a panic button. No public quantum computer can break Ethereum’s signatures today, and analysts do not expect the proposal itself to move the ETH price.
It is a plumbing change: make the front door wide enough for larger keys, then install a lock that can be thrown once and never reopened the other way. If the community accepts it, staking’s entry path will be ready when a real post-quantum standard is chosen. If it stalls, the network will still be arguing about key sizes when those standards arrive.