Ethereum co-founder Vitalik Buterin has outlined a potentially significant shift in how Ethereum transactions are structured and validated, pointing to recent progress in account abstraction, state design, and cryptographic proof aggregation.
In a post shared on September 6, 2026, Buterin argued that research around EIP-8141, alternative state models, keyed nonces, and recursive STARK-based mempools has produced a more explicit conceptual model of transactions. The central idea is to separate a transaction into two components: “actions” and “dependencies.”
An action is the effect a transaction has, while dependencies are the conditions or information that must be proven before the action can execute. Buterin said this distinction may sound conceptual, but it matters because dependencies have properties that make them especially suitable for optimization.
Many dependencies can be checked in parallel, some can be examined by the mempool before a transaction reaches a block, and others do not require access to Ethereum state at all. These so-called “pure” dependencies could potentially be checked once by the mempool rather than repeatedly executed by every validator. In the long term, Buterin suggested, large collections of such checks could eventually be replaced by a single STARK proving that they were all performed correctly.
This separation of validation and execution could allow Ethereum developers to optimize each component in fundamentally different ways, potentially improving scalability and efficiency across the network.
