Flashbots scientist: Encrypted mempools are not yet economically feasible; programmable privacy may be the better path

Gate News message, April 2, Flashbots Senior Research Scientist and visiting scholar at Imperial College London Jonathan Passerat-Palmbach presented a systematic account of the technical evolution and real-world limitations of the Encrypted Mempool at EthCC[9]. He noted that from early committee-based threshold encryption schemes, to Batched Threshold Encryption, and then to the Beast-Mev approach that combines silent setup with batched decryption, academia has proposed multiple construction paths. However, no current scheme can simultaneously satisfy all the protocol-level deployment requirements, such as silent setup, non-interactive decryption, small keys, and small ciphertexts. Therefore, the cryptographic component of an encrypted mempool cannot yet be directly embedded into the Ethereum protocol.

More importantly, the key issue lies in the economic layer. Encrypted mempools use a Blind Ordering mechanism that orders transactions while they are encrypted, which leads to three problems: 1. Searchers will still submit large numbers of blindly targeted arbitrage bundles, resulting in on-chain garbage transactions; 2. Reduced MEV subsidies will make validators in remote geographic locations lose competitiveness, thereby further exacerbating centralization; 3. Users’ execution quality declines, transaction price deviations increase significantly, and the probability of rollback rises.

Jonathan said that the timing design of decrypting after execution prevents the auction mechanism from running effectively, and that auctions are the best way to handle MEV incentives. He proposed programmable privacy as an alternative direction, enabling a middle state between fully encrypted and fully public, which can be achieved today via TEE and in the future could incorporate more advanced schemes such as FHE. He concluded that privacy cannot exist independently of economic incentives; the right approach is to continue experimenting outside the protocol rather than rushing to embed it.