Vitalik Buterin’s wallet got hit by jaredfromsubway.eth last week. A sandwich attack. The same MEV bot that’s been bleeding retail across Uniswap for years. The same bot that exploits public mempool visibility. The same bot that Ethereum’s core developers now claim they can stop with a new set of EIPs.
They can’t. Not yet. Not for a long time.
On August 19, the “Encrypt the Mempool” call laid out the roadmap: EIP-8184 (LUCID), EIP-8105, and FOCIL. The pitch is simple—hide the transaction content before block inclusion, kill the information asymmetry that MEV bots feed on. The reality is a cryptographic dead end. No known construction satisfies all requirements. The proposed workaround? Trust a third party. Again.
I count the cracks before the dam breaks. This dam has more cracks than concrete.
Context: The Mempool as a Battlefield
Every pending Ethereum transaction sits in the public mempool—visible to miners, validators, and anyone running a node. MEV bots scan this pool for arbitrage or liquidation opportunities, then front-run or sandwich the original transaction. The result: users pay more slippage, bots capture the spread. Since 2020, MEV extraction has exceeded $1.5 billion on Ethereum alone.
Current mitigations are off-chain: private relays, encrypted transaction channels, or trusted sequencers. They work—but they introduce a new central point of control. The relay operator sees everything. If the operator goes rogue or gets compromised, the user’s privacy is null.
The protocol-layer fix promises to eliminate the need for trust. Encrypt the mempool so that no one—not even validators—can see the transaction until it’s committed. Then reveal the key after inclusion. That’s LUCID’s core idea.
But the devil is in the cryptographic primitives, and the devil is winning.
Core: The Technical Guts That Bleed
Let’s dissect LUCID (EIP-8184). It’s a commit-reveal scheme. The sender submits a sealed transaction (encrypted) along with a deposit. The block builder includes it blind. After the block is built, the sender releases the decryption key. The transaction is then executed. If the key never arrives, the builder can’t execute the transaction, but the sender loses the deposit.
Sounds clean. But the decryption key release is not enforced by the protocol. It’s handled by the sender or a third-party key publisher. That’s a trust assumption. The same trust assumption that private relays require.
EIP-8184’s authors openly admit: “No known cryptographic construction meets all requirements simultaneously.” The list of unmet requirements is brutal:
- Small public keys (non-interactive)
- No trusted setup
- Practical ciphertext size
- Strong chosen-ciphertext security
- A credible path to post-quantum security
Post-quantum is the real time bomb. Q-Day could arrive by 2029. If Ethereum’s encrypted mempool launches in 2027 with FOCIL (the Hegotá upgrade), it will have a two-year window before quantum risk becomes critical. And there’s no post-quantum encryption scheme that fits Ethereum’s performance constraints today.
Then there’s EIP-8105, which proposes a “directed trust graph”. Senders register which providers they trust to relay encrypted transactions. Trust is explicit, but penalties for misbehavior are off-chain. No protocol slashing. No forced compliance. If a provider colludes with a bot, the user has no recourse except reputation damage.
FOCIL (EIP-7805) adds an inclusion list pipeline. Multiple validators can force a block builder to include certain transactions. This prevents censorship but doesn’t solve the encryption problem. It’s a separate mechanism bolted on.
Together, these proposals form a fragile stack. The encryption layer is incomplete. The trust layer is outsourced. The inclusion layer is delayed until 2027. The bottom line: Ethereum’s encrypted mempool is a prototype, not a product.
Contrarian: The Real Goal Is Not Privacy—It’s Raising the Cost of Attack
Most coverage frames this as Ethereum fighting MEV. I see a different motive. The design choices—like limiting the encrypted block top segment to 1/8 of the gas limit and requiring a reservation fee—suggest the real goal is to make MEV attacks economically unviable, not technically impossible.
A bot that wants to sandwich a transaction must pay a reservation fee, commit to a block, and hope the decryption key arrives. If the key is delayed, the bot loses the fee. The cost of attack increases. The frequency of attacks decreases. But the underlying vulnerability remains: if the bot can predict the transaction content through other means (e.g., off-chain information), it can still front-run.
This is a cat-and-mouse game, not a permanent fix.
Moreover, the encrypted mempool introduces new centralization vectors. The key publisher becomes a single point of failure. If the publisher is a regulated entity, it may be forced to disclose keys. If it’s a DAO, governance attacks become possible. The “trustless” Ethereum is swapping one set of trusted parties for another.
Private relay proponents argue that their solutions are already production-ready. They’re right. The encrypted mempool is years away. The battle is not about eliminating MEV—it’s about controlling who captures it.
Takeaway: The Timeline Is 2027—At Best
FOCIL is a priority for Hegotá, which is the next consensus layer upgrade after Pectra. That puts the earliest possible deployment at 2027. LUCID and EIP-8105 are not even scheduled for a testnet. They’re research concepts.
Meanwhile, the bots are running. jaredfromsubway.eth generated over $30 million in profit in 2023 alone. Retail users continue to lose slippage. The only live defense is private relays, and they’re centralized.
I’ve been auditing smart contracts and trading crypto since 2017. I’ve seen ICOs promise trustless systems and deliver backdoors. I’ve seen DeFi protocols collapse because of economic design flaws, not code bugs. The encrypted mempool is another case where the promise exceeds the cryptographic reality.
Survival is the only alpha that compounds. Right now, the smart move is to use private relays, monitor mempool congestion, and avoid trades during high MEV activity. The protocol fix is a mirage on the horizon.
Liquidity is just borrowed time with a premium. Ethereum’s encrypted mempool is borrowing time from cryptography that hasn’t been invented yet. The clock is ticking.
