Vitalik Buterin Proposes A New Cryptography Model For Ethereum

Thanks to EIP-8288, Ethereum cryptography co-written by Vitalik Buterin and Thomas Coratger could become less costly. This text will have to group post-quantum signatures and STARK proofs, however it remains in draft form.

One proof would replace thousands of verifications

Signatures that resist quantum computers remain too heavy for Ethereum. Thus, the studied models occupy about 2 to 3 KB and require between 150,000 and 200,000 gas units per verification.

STARK proofs pose an even greater difficulty. Indeed, their size goes beyond 128 KB and can approach 512 KB if their creation must remain fast. Direct verification on Ethereum would then cost several million gas units.

The EIP-8288 therefore offers to replace each complete proof with a lighter cryptographic dependency. Its operation could rely on four main phases :

• The transaction declares a claim, such as the validity of a signature ;
• A user transmits the corresponding proof to the mempool nodes ;
• The nodes group the proofs into a common recursive STARK ;
• The block contains a single aggregated proof and a 96-byte tag per dependency.

Mempool nodes would produce a new aggregate every second. Then, the block builder could gather these packets into a final proof, recorded in the block header.

The volume related to STARKs would remain close to 256 KB per aggregation interval, regardless of the number of proofs grouped. However, transactions could continue to circulate separately on the network.

The operation therefore does not remove the cryptographic cost. It shifts much of the work off the blockchain, then requires the network to verify a single common proof.

Ethereum cryptography would gain four uses

In this architecture, Vitalik Buterin first sees a way to reduce the cost of post-quantum signatures. Ethereum would thus integrate mechanisms based on leanSPHINCS without storing each complete signature in a block.

Privacy protocols are concerned with the second use. Thus, users would prove the validity of an operation without revealing all the information that composes it. Grouping would then reduce the fees related to STARK proofs.

EIP-8288 could also allow the integration of new signature or proof systems. A user would wrap a cryptographic mechanism in a compatible STARK without imposing its full format on the protocol.

Buterin finally mentions a “private account abstraction”. A user would modify the owner of various on-chain positions in a transaction, without publicly revealing the assets moved.

Such a feature would facilitate wallet recovery or key rotation if compromised. However, it could not immediately protect all Ethereum accounts against quantum computers.

Wallets and applications would have to adopt the new signatures. EIP-8288 only provides the infrastructure useful to verify them at lower cost.

However, the system also requires a common environment for writing and checking proofs. According to the provided information, Buterin presents RISC-V as the main candidate to fulfill this role.

No deployment is scheduled yet

On June 3, 2026, Vitalik Buterin and Thomas Coratger created EIP-8288. This proposal notably depends on EIP-8141, which introduces transactions composed of various execution frames.

The document retains draft status. Its aggregated verification key remains to be defined, while the section for reference implementation still indicates “to be determined”.

Its activation could also change consensus rules. Old nodes would reject transactions and blocks using this new format. Ethereum would thus integrate EIP-8288 in the network upgrade.

Buterin hopes for inclusion in “I-star”, which he describes as “the fork after Hegota”. This statement represents a technical preference, not a confirmed deployment schedule.

Many risks remain to be resolved. Security will depend on the reliability of Lean Ethereum circuits, the hash function, and STARK recursion. Therefore, developers will also have to limit denial-of-service attacks.

The EIP already projects a maximum of 16 leanSPHINCS signatures and one leanSTARK proof per transaction. These parameters may still evolve during discussions and tests.

EIP 8288 would thus open a new stage for Ethereum cryptography. Its adoption will nevertheless require a functional implementation, audits, and developer agreement before any activation.