According to Foresight News , Vitalik Buterin stated that "Keyed Nonces" not only provide stronger protocol-level support for on-chain privacy solutions, but may also be an important direction for Ethereum's future state expansion. By creating dedicated storage types for specific use cases, extreme scalability can be achieved while maintaining the protocol's decentralization.
Taking nullifiers in privacy transactions as an example: if on-chain privacy transactions maintain a TPS of 2000, approximately 500 billion nullifiers will be generated after 8 years. Since nullifiers cannot be pruned, placing them all in the existing general state would put enormous pressure on node storage and decentralization. Dedicated nullifier storage can leverage its highly restricted usage (only for validity checks, and transactions must explicitly provide nullifier IDs) to achieve more decentralized processing solutions, such as sharded storage (nodes only need to hold a small number of shards and maintain connections with honest nodes in each shard) and Bloom filter optimization (which can significantly reduce verification overhead to approximately 8 bits per nullifier).
If we continue to rely on the existing state, builders may need to download tens of terabytes of full data, and privacy-conscious users will not be able to use mechanisms like FOCIL efficiently. Therefore, most use cases should be migrated to more specialized and lower-cost storage types, while retaining the complete dynamic state for applications requiring high flexibility, such as DeFi, thereby achieving a better balance between scalability and decentralization.
