Layer-1 blockchain Aptos has officially rolled out an optional new feature designed to safeguard traders, institutions, and emerging decentralized applications from the financial hazards of maximal extractable value, commonly known as MEV. The network announced on Monday that its Encrypted Mempool feature is now live on the mainnet, offering users an opt-in mechanism to keep their pending transactions shielded from prying eyes until block ordering has been firmly finalized.
The introduction of the tool addresses one of the most persistent and costly vectors of value extraction in public blockchains. According to Aptos, the infrastructure was engineered specifically with multiple user segments in mind, catering to institutional entities executing large-scale token transfers, everyday traders looking to evade predatory MEV strategies, builders of decentralized applications, and autonomous artificial intelligence agents operating on-chain.
For years, the transparent nature of public ledgers has left transaction details exposed the moment they are broadcast to a network. When a user submits a transaction on any major blockchain today, anyone watching the network can view the specific details before execution takes place. Observers and automated entities can see the function being called, the exact assets and amounts involved, and the specific smart contracts being interacted with.
Bots programmed to monitor these broadcast channels exploit this radical visibility to extract profit at the expense of unsuspecting users. These automated scripts use the information to trade directly against incoming orders or copy profitable strategies in real time. This dynamic most frequently manifests as frontrunning, where a bot pays a higher gas fee to jump ahead of a pending trade, or sandwich attacks, where a bot buys an asset just before a large purchase drives the price up, and then sells it immediately afterward for a guaranteed profit.
The financial toll of these predatory practices across the broader blockchain ecosystem is substantial. Recent market data highlights the magnitude of the problem on competing layer-1 and layer-0 networks. On Ethereum, sandwich attacks cost traders approximately $60 million over the 12-month period leading up to October 2025, according to an analysis of EigenPhi data. Meanwhile, the Solana ecosystem experienced a similar onslaught, where sandwich bots generated $29.7 million in illicit or extractive value over an 11-month examination window conducted by Bitquery. That investigation identified nearly 6.9 million distinct attacks that successfully impacted roughly 2.2 million unique wallets. By introducing the Encrypted Mempool, Aptos is aiming to give its community the cryptographic tools necessary to neutralize these automated threats before they can impact execution prices.
How Encrypted Mempool Works
The operational mechanics behind the newly deployed feature rely on a multi-stage cryptographic process designed to blind validators and network observers to transaction details during the crucial window between submission and ordering. When a user chooses to opt in to the protection, a compatible wallet or decentralized application encrypts the contents of the transaction locally on the user’s device before transmitting the data packet to the Aptos network.
Once encrypted, sensitive parameters such as the specific assets being traded and the exact amounts involved remain completely hidden from public view. However, vital routing and validation metadata, including the sender’s wallet address and the designated gas fee, remain visible to ensure the network can process the transmission appropriately.
The encrypted transaction then enters the network mempool, which functions as the waiting room where pending transactions queue up prior to being packaged into a new block. During this phase, network validators coordinate to reach consensus on the sequence and ordering of the incoming transactions while remaining entirely blind to the actual payloads inside them.
This development aligns with broader strategic pushes within the ecosystem, arriving shortly after the Aptos Foundation and Aptos Labs committed a substantial $50 million to bolster AI agent infrastructure, recognizing the growing necessity for secure, automated transaction environments.
Once the consensus mechanism finalizes the sequence of the blocks, the network validators collectively decrypt the transactions. Only after this decryption phase are the orders executed and permanently recorded on-chain. Crucially, the cryptographic architecture ensures that no single validator can decrypt a pending transaction on its own, preventing internal censorship or premature information leaks.
By keeping transaction details concealed until their definitive ordering is locked in, the system effectively prevents automated bots from reading pending orders in the mempool. Without the ability to inspect the payload, predatory programs cannot insert trades ahead of or around legitimate user orders to extract value at their expense.
This advanced level of protection does, however, introduce minor operational trade-offs in terms of cost and processing latency. According to official developer documentation released by Aptos, transactions routed through the encrypted mempool carry a mandatory minimum gas price set at twice the network’s standard base minimum. Furthermore, internal benchmark tests conducted by the development team across a network of 50 nodes demonstrated that the cryptographic overhead adds roughly 20 milliseconds of latency per block when compared to the existing, unencrypted consensus implementation.
With this deployment, Aptos joins a select group of blockchain networks offering built-in or native defenses against harmful MEV extraction. Other networks have tackled the problem through varied architectural approaches. Gnosis Chain introduced an encrypted mempool utilizing Shutter’s threshold encryption framework in July 2024, while Ethereum relies heavily on services like Flashbots Protect, which routes user transactions through private channels directly to trusted block builders rather than a public pool.
Aptos emphasizes that its primary architectural distinction lies in its utilization of its existing validator set. Instead of relying on a separate, dedicated decryption committee or third-party trusted builders, the native Aptos validators handle both the sequencing and the collaborative decryption of the transactions directly, maintaining a decentralized trust model.
