When Vitalik Buterin's wallet gets hit by Jaredfromsubway.eth, the MEV problem becomes personal. The attack wasn't a sophisticated exploit—just a frontrunning bot that spotted a pending transaction and jumped ahead. But it happened to the co-founder of Ethereum. That irony frames the urgency behind a new wave of proposals: LUCID, EIP-8184, EIP-8105, and FOCIL. They promise to hide your trades from bots before they can strike. But after two decades of auditing protocols, I've learned that promises are cheap. Let's look at the code—or rather, the lack of it.
Context: The Mempool's Open Wound
The public mempool is Ethereum's Achilles' heel. Every pending transaction is visible to anyone running a node. Bots monitor this stream, frontrun trades, sandwich users, and extract millions in MEV. Current solutions rely on private relays—trusted intermediaries that sequester transactions away from public view. But trust is a single point of failure. The August 19 "Encrypt the Mempool" call signaled a shift: the Ethereum research community is now pushing for protocol-level encryption. The proposals are LUCID (EIP-8184), a commit-reveal scheme; EIP-8105, a directed trust graph; and FOCIL (EIP-7805), an inclusion list mechanism. They are bundled into the Hegotá upgrade consensus layer priorities, slated for 2027. That timeline already raises red flags.
Core: Where the Cryptographic Bone Breaks
LUCID's design is elegant on paper. A block builder submits a sealed transaction blob without knowing its contents. After the sealed blob is committed, a decryption key is released—either by the sender or a third-party key publisher. The builder then reveals the transactions. The problem? The author of EIP-8184 admits: "No known cryptographic construction meets all requirements at Ethereum scale." Those requirements include small public keys, non-interactive decryption, no trusted setup, practical ciphertext size, strong chosen-ciphertext security, and a credible path to post-quantum security. That's a laundry list of unsolved problems. I've spent years reverse-engineering ICO code and DeFi arbitrage mechanics, and I can tell you when a proposal is still in the PowerPoint stage. LUCID is a PowerPoint.
EIP-8105 tries a different angle: a directed trust graph where registered providers can indicate which other providers they trust. Trust, incentives, reliability, and penalties are all left outside the consensus layer. That means no on-chain slashing for misbehavior. From my experience auditing governance contracts after the Terra collapse, I know that off-chain penalties are rarely enforced. The system relies on reputation—which is just another form of centralization. FOCIL adds a layer of multiple validators who can force the inclusion of specific transactions, but it's still a pipeline, not a solved puzzle.
Contrarian: The Real Goal Isn't Elimination—It's Cost
Most coverage reads these proposals as a silver bullet against MEV. I'm not buying it. The encryption schemes are designed to raise the cost of attack, not eliminate it. LUCID limits the encrypted block top section to one-eighth of the block gas limit and uses a reserve fee. Most of that fee is refunded upon successful decryption, but failure can cost the sender the entire reserve. This is a deposit-and-slash mechanism, not a privacy guarantee. The real insight is that Ethereum is willing to accept a trust shift: from trusting bots not to snoop, to trusting key publishers not to collude. That's a marginal improvement, but it's not a revolution.
Furthermore, the cryptographic bottleneck is unlikely to break before the Hegotá upgrade. Post-quantum security is a stated requirement, yet no existing construction works. The timeline for "Q-Day" is estimated to be as early as 2029. If Ethereum's encrypted mempool can't even reach a testnet before that, it will be obsolete before it launches. Private relays, for all their flaws, are production-ready today. They are the de facto solution, and they will remain so for the next 3-5 years.
Takeaway: Code Executes, Hype Crashes
The encrypted mempool is a necessary direction, but the cryptographic foundations are not yet laid. I've seen too many projects promise trustless solutions that end up shifting trust to a new middleman. LUCID's key publisher is that middleman. EIP-8105's trust graph is that middleman. FOCIL's validator set is that middleman. The only way to truly eliminate MEV is to eliminate the information asymmetry at the protocol level—and we're decades away from that. For now, the best defense is a private relay and a healthy dose of skepticism. Logic prevails where hype fails to compute.
