Hegota as final recognizable fork

Ethereum co-founder Vitalik Buterin has labeled the Hegota fork, scheduled for 2027, as the chain's last "normal" upgrade, positioning what follows as a "cryptographic world computer" in a Saturday post.

In the essay, Buterin said the network after Hegota is called a blockchain "to a large extent for historical reasons." He characterized it as a "hybrid construction" merging Satoshi Nakamoto's consensus design with cryptographic tooling that "did not even exist" in 2009.

The post functions as a plain-language companion to the Strawmap, the Ethereum Foundation's draft schedule of seven forks through 2029 that researcher Justin Drake published in February.

What changes after Hegota?

Three core properties are being replaced in the architecture Buterin outlined:

• Verification moves from re-executing every block to sampling data via PeerDAS and checking a SNARK
• Consensus shifts to a "much better optimized" proof-of-stake model
• Block construction transitions from one miner to multiple parties building together

Finality would compress to 8 to 32 seconds from roughly 200 seconds for 12 confirmations today, Buterin wrote. Slots would run 4 to 8 seconds on the near-term design before later sliding to two seconds on the Strawmap timeline.

FOCIL governs censorship resistance. Each validator on a committee publishes a list of pending transactions the next block must include. The fork-choice rule ignores any block that omits those transactions. Buterin described the outcome as "guaranteed real-time transaction inclusion."

How does privacy fit?

Privacy arrives in uneven tiers, according to the post. General-purpose computation remains expensive and only partly private, while purpose-built applications receive "very strong privacy" from zero-knowledge proofs and private account abstraction. Onion routing and mixnets cover the network layer, an effort the Ethereum Foundation resourced in October 2025 through a dedicated privacy cluster.

Developers will feel the shift in gas costs, Buterin warned. The same workload costs more "if you shove it all into one inscrutable serially-executed transaction," he wrote, and less when split into parallelizable dependencies and stripped of redundant operations before reaching a block.

Why does decentralization still pay off?

Buterin challenged the long-running view of decentralization as pure overhead. He argued it enables parallel data storage and parallel computation inside the mempool, letting metadata be hidden in ways centralized systems cannot match.

That vision, he wrote, was the "mid-2010s dream," and it failed because no efficient method existed to verify that split work had been done correctly. Randomly sampled committees proved "complicated to set up and expensive," adding latency without a recovery path if they failed. Modern cryptography resolves the trade-off, Buterin said, with overhead "decreasing month by month."

What's still unresolved?

Further out sits indistinguishability obfuscation, which Buterin called the privacy-and-generality end state. He stressed that the current roadmap does not depend on it: "all of the conclusions in this post will apply long before any of that becomes available."

The harder near-term problem is operational. Making zero-knowledge proofs fast and safe enough amounts to "encapsulated complexity" already under AI-assisted optimization. Managing and parallelizing access to very large state remains "more difficult, and systemically complex."

The Strawmap charts the engineering dependencies fork by fork. Buterin's essay points at the destination: a network rebuilt component by component in what he previously called a "ship of Theseus" exercise, with finality sliding from roughly 16 minutes today toward single-digit seconds by the end of the roadmap.