Aptos has activated its Encrypted Mempool on mainnet, introducing protocol-level transaction encryption designed to reduce front-running and sandwich attacks. The feature allows users to conceal transaction instructions before submission, preventing validators and transaction-monitoring bots from inspecting their contents while determining block order.
The official announcement, published October 12, confirms that encryption is optional for individual transactions and operates within Aptos’ existing validator infrastructure. Unlike approaches that depend on external committees or specialized execution environments, the implementation uses the network’s established consensus participants to coordinate transaction decryption.
The deployment follows the activation of Block-STM v2 as Aptos’ default execution engine, another infrastructure upgrade supporting the network’s broader performance strategy. Together, the developments expand Aptos’ transaction-processing capabilities while addressing different aspects of blockchain execution and security.
How Aptos Encrypted Mempool Protects Transactions
Under the new system, transaction payloads are encrypted directly on the user’s device before entering the mempool. Validators can order encrypted transactions without knowing which functions they invoke, what assets they involve or which parameters they contain.
The mechanism relies on threshold encryption integrated into AptosBFT, the network’s consensus protocol. Validators collectively generate an encryption key for each epoch, while transaction decryption requires sufficient stake-weighted participation.
According to the AIP-144 technical specification, encrypted transactions are decrypted after block consensus finishes but before execution begins. This sequencing prevents validators from using knowledge of transaction payloads to rearrange pending transactions for their own benefit during the ordering process.
The protection has defined limitations. Sender addresses and gas-fee information remain visible, while transaction contents become publicly accessible after finalization. Consequently, encrypted mempool functionality does not provide permanent transaction confidentiality or eliminate every possible form of MEV.
Aptos also designed the system around batched threshold encryption to reduce computational and communication overhead. Its technical proposal describes a launch performance target exceeding 1,000 transactions per second, although that target should not be interpreted as independently verified production throughput.
Petra Integration Expands Access to Transaction Privacy
The encrypted mempool is already integrated into Petra, where users can enable encryption directly during transaction signing. Developers can also access the functionality through version 7.0.0 or later of the Aptos TypeScript SDK without modifying existing Move smart contracts.
Aptos identified decentralized trading as an important application, particularly where visible orders can expose users to predatory execution strategies. Decibel, a perpetual futures exchange being developed within the ecosystem, intends to use encrypted transactions to protect order information before execution.
The rollout also complements confidential APT functionality introduced through Petra. However, the mechanisms serve different purposes: confidential transfers protect selected financial information, while encrypted mempools conceal transaction instructions during ordering.
Interest in this architecture extends beyond Aptos. Within Ethereum, initiatives involving the Erigon execution client are pursuing comparable protections against transaction-ordering exploitation, although their implementations and development stages differ.
For Aptos, native encryption shifts transaction protection from optional external services into the blockchain’s consensus architecture. Its effectiveness against harmful MEV will depend on how consistently applications and users adopt encrypted submissions, alongside the security of the threshold decryption process.
The mainnet deployment establishes the technical capability, but Aptos has not disclosed sufficient post-launch data to quantify any reduction in sandwich attacks, front-running losses or other forms of extractable value.
Jack Reynolds is SatoshiPick’s infrastructure mind. Based in Denmark, he follows Bitcoin, Ethereum, Layer 1 networks, stablecoins and DeFi with a practical question always in the background: does this actually make crypto work better?
His articles focus on the systems beneath the headlines: settlement rails, protocol upgrades, stablecoin usage, DeFi coordination and the regulatory limits around them. Jack avoids turning technical coverage into a maze. He explains what is live, what is still experimental and why the difference matters.