Vitalik Buterin laid out a phased plan on March 2 to reduce the outsized control that a small number of specialized "block builders" hold over which Ethereum transactions get included in blocks, and in what order. The proposal comes ahead of Glamsterdam, Ethereum's next hard fork targeting a first-half 2026 launch, and sketches a longer-term research agenda extending well beyond it.

Why Builder Centralization Is a Problem Now

Since Ethereum switched to proof-of-stake in 2022, over 90% of its blocks have been produced through MEV-Boost, an off-chain software layer that lets specialized builders compete in an auction to fill each block. The system has created a concentrated market: two to three builders now control an estimated 80 to 95% of all Ethereum blocks, according to data from Blocknative and blockchain analytics firm emergentmind. Among the dominant players as of late 2024 are beaverbuild.org, Titan Builder, and rsync-builder.

Those builders profit largely from MEV, short for "maximal extractable value." The most harmful form, known as sandwich attacks, involves bots spotting a pending user trade in the public mempool, placing orders on both sides of it, and pocketing the difference. In 2025, sandwich attacks accounted for $289.76 million, or roughly 51.6% of the $561.92 million in total MEV transaction volume tracked on Ethereum. The average DeFi user loses between 0.5% and 3% of their transaction value to these tactics.

The concentration also creates a censorship risk. After U.S. regulators sanctioned the crypto mixing service Tornado Cash in 2022, five of Ethereum's six largest block builders began filtering out associated transactions. At the peak of that episode, nearly 99% of relay-mediated blocks were excluding OFAC-blacklisted transactions.

Buterin acknowledged that Glamsterdam's headline features include enshrined Proposer-Builder Separation (ePBS, EIP-7732), which brings the builder auction onto Ethereum's base layer and removes the need for trusted relay intermediaries, as well as Block-level Access Lists (BALs, EIP-7928). But he noted that ePBS "doesn't resolve the risk that a small number of builders could still dominate and censor transactions."

What Buterin Is Proposing

The next upgrade after Glamsterdam, called Hegota and targeting the second half of 2026, will introduce FOCIL (Fork-Choice Enforced Inclusion Lists, EIP-7805). The mechanism works by randomly selecting 16 validators each slot, requiring each to nominate a set of transactions for inclusion. Blocks that omit those transactions are rejected by attesters and cannot become canonical.

Buterin has described it as enabling "censorship-resistant rapid inclusion of any transaction," with confirmed transactions landing within one to two slots even under adversarial conditions. His enthusiasm for the combined FOCIL and EIP-8141 approach was plain: he remarked that "Ethereum is going hard" on the problem.

Buterin is also floating a more aggressive extension he calls "big FOCIL." In this model, inclusion lists would grow large enough to cover most or all transactions in a block, leaving builders responsible only for ordering a narrow slice of MEV-relevant activity such as DEX arbitrage.

It would, in effect, reduce the block builder's role to a minor function rather than a gatekeeping one. Looking further out, Buterin pointed to encrypted mempools, where transactions remain hidden from observers until after block inclusion, eliminating the window that sandwich bots currently exploit. He also flagged network-layer privacy, including Tor-based routing and Ethereum-specific mixnets, as an underexamined attack surface.

Both remain active research areas without confirmed implementation timelines.

Not everyone is supportive of the direction. Ameen Soleimani, founder of SpankChain and an early Ethereum contributor, warned that FOCIL's forced inclusion design could create "unforeseen ramifications" for validators when it comes to transactions involving sanctioned addresses, a concern the community has not fully resolved.

Real Costs for Users in Emerging Markets

These protocol changes carry direct relevance for the regions where Ethereum and DeFi are growing fastest. India, home to roughly 150 million crypto users and the largest crypto-owning population globally, saw South Asia record 80% growth in crypto adoption in the first half of 2025. Sub-Saharan Africa received $205 billion in on-chain crypto value between July 2024 and June 2025, a 52% year-over-year increase. Nigeria alone accounted for $92.1 billion of that figure. In Nigeria, where the naira has faced significant devaluation and access to dollar stablecoins through official channels has been restricted, decentralized and censorship-resistant stablecoin access has become functionally significant for ordinary users.

In both regions, a significant share of on-chain activity comes from ordinary users transacting for remittances, savings, and payments rather than institutional trading. A 0.5% to 3% MEV tax on a cross-border stablecoin transfer is a meaningful loss at those income levels.

An encrypted mempool would remove the attack vector entirely. FOCIL's censorship resistance is equally relevant in jurisdictions where government pressure on financial intermediaries is a realistic threat. For developers building smart account applications in Lagos or Bangalore, FOCIL's guaranteed inclusion also means that gas-sponsored and batched transactions under account abstraction standards such as ERC-4337 and the proposed EIP-8141 will become more reliable once Hegota ships. Buterin has specifically cited smart wallet transactions under EIP-8141 as explicit FOCIL beneficiaries. One practical caveat for these regions: FOCIL does impose modest but real network requirements on solo stakers, a consideration that carries weight where internet infrastructure is variable.

What Comes Next

FOCIL was formally added to the Hegota roadmap at the February 19 All Core Devs meeting, giving it a clear path to production. Glamsterdam is scheduled for mainnet in the first half of 2026, with Hegota targeting the second half. The longer-term proposals, big FOCIL and encrypted mempools, remain at the research stage. Buterin's framing suggests Ethereum is treating builder centralization as a structural problem requiring layered solutions across multiple upgrade cycles, not a single fix.