Flashnet aims to mitigate metadata leakage (IP, timing, network observability) during submission via anonymous broadcast. Shielded execution environments like Aztec and Railgun protect intent, state changes, and MEV exposure during processing. Yet a key gap persists: inclusion. Even anonymized and shielded transactions can be silently censored by builders in a PBS environment where block production is concentrated. FOCIL (EIP-7805) and Frame Transactions via EIP-8141 (building on account abstraction primitives) propose to make inclusion a protocol-enforced guarantee for privacy protocols and advanced accounts, assuming activation in the upcoming Hegota fork.
This matters because privacy in DeFi is only credible if transactions reliably reach on-chain settlement without depending on builder goodwill. The design shifts Ethereum toward structural censorship resistance at the base layer.
The Inclusion Gap in the Privacy Pipeline
Prior articles in this series highlighted:
- Centralized intermediaries and fiat rails as primary barriers to private payments.
- Flashnet's anonymous broadcast closing submission-layer leaks.
The missing piece is inclusion: PBS enables builders to drop transactions without consequence, reintroducing censorship risk precisely where privacy protocols operate. Without enforced inclusion, the pipeline remains fragile—shielded logic can execute privately but never confirm.
FOCIL: Enforcing Inclusion via Fork-Choice
FOCIL (EIP-7805), now SFI’d as the consensus-layer headliner for Hegota, introduces fork-choice-enforced inclusion lists:
- Each slot assigns the proposer plus 16 randomly selected includers (validators).
- A bounded inclusion list must be processed; qualifying transactions are forced into the block.
- Initial payload ~8 kB, with upgrade path planned.
Properties that strengthen censorship resistance:
- Randomization forces coordination across 17 actors per slot for suppression.
- Fork-choice rule penalizes bypass.
- Expected inclusion within 1–2 slots under standard conditions.
It preserves ePBS MEV auctions for final block ordering while limiting single-proposer veto power.
Frame Transactions (EIP-8141) as the AA Complement
EIP-8141 introduces Frame Transactions, abstracting validation, execution, and gas payment to make advanced account behaviors native first-class senders. Privacy protocols (via paymasters or potential 2D nonces), multisigs, quantum-resistant signatures, and sponsored flows submit directly—no wrappers or centralized relays as chokepoints. This ensures shielded or sponsored operations are indistinguishable from standard txs at the inclusion layer, preventing type-based filtering.
Vitalik Buterin discussed this synergy in a February 19, 2026 post responding to FOCIL's prioritization for Hegota:
He further noted the design "ensures that even if literally 100% of all slots get sold off via proposer-builder separation to a hostile actor that refuses to connect to public mempools, discriminates against certain applications, or is otherwise abusive, all transactions can still get quickly included.""There is also an important synergy between FOCIL and AA (EIP-8141, which is based on 7701): 8141 makes not just smart accounts [...] first-class citizens, it also can do the same for privacy protocols [...] Hence, with FOCIL and 8141 together, anything, including smart wallet txs, gas sponsored txs, and even privacy protocol txs, can be included onchain through one of 17 different actors [...] almost certainly within 1-2 slots, of any such tx, even in an adversarial environment."
Vitalik Buterin on X
This preserves ePBS for MEV last-look while materially weakening single-proposer veto power—critical for privacy flows that can't afford silent drops.
The Emerging Privacy Pipeline (February 2026)
| Layer | Protects | Leading Tools | FOCIL + EIP-8141 |
|---|---|---|---|
| Submission | Sender metadata, IP, timing | Flashnet anonymous broadcast | Enables routing to FOCIL inclusion path |
| Inclusion | Guaranteed landing against adversarial builders | Public mempool + PBS (best-effort) | Forced inclusion via 17 random actors; ~1–2 slot bound |
| Execution | Intent, state, MEV leakage | Aztec, Railgun, Secret, Oasis + encrypted mempools | Native support for shielded / advanced account ops |
| Settlement | On-chain linkability | Shielded pools, ZK proofs, stealth addresses | Reinforces unlinkable anonymity sets |
| Merchant / Real World | Identity linkage at fiat boundary | Nascent crypto-native rails | Unresolved; supports private flows only if rails evolve |
The pipeline becomes contiguous from broadcast to inclusion, reducing dependence on cooperative builders.
Second-Order Effects
- Raises censorship costs significantly: suppression requires sustained multi-actor coordination rather than unilateral action.
- Enables reliable private agentic DeFi: automated strategies (vault rebalances, intent solvers, private liquidations) face lower execution uncertainty.
- Embeds adversarial assumptions deeper into protocol design: anonymous submission + enforced inclusion + native AA moves Ethereum toward explicit censorship resistance, prioritizing resilience over cooperative defaults.
Remaining Constraints
These upgrades harden on-chain layers but leave economic centralization in block building, fiat-bound merchant rails, and regulatory pressures on wallets/infrastructure untouched. Private transactions still need viable off-chain endpoints to matter economically.
Areas Requiring Scrutiny
Early client prototypes exist, but adversarial simulations at scale are still limited.
- Inclusion lists may function mainly as a deterrent (rare invocation could suffice via credibility).
- 8 kB initial payload may limit complex shielded or frame transactions.
- Includer incentives and potential alignment/collusion with dominant builders.
- Economic penalties that could undermine technical guarantees.
Conclusion
Submission anonymity via Flashnet, shielded execution via Aztec/Railgun, and now proposed enforced inclusion via FOCIL + EIP-8141 bring the privacy pipeline closer to base-layer completeness. Private transactions become harder to suppress structurally. The frontier shifts from architectural gaps to economic, regulatory, and real-world integration challenges.