Ethereum co-founder Vitalik Buterin said Hegota, expected to ship next year, will likely be the network’s last “normal” fork, with later upgrades changing in character and focusing on proof systems and other advanced cryptography.
In a blog post published Sunday and shared on X, titled The cryptographic world computer, Buterin wrote: “Hegota is likely to be the network’s last normal fork.” He added that “everything after that involves recursive STARKs, automated formal verification, highly optimized consensus algorithms, and making it all quantum-safe.”
A hybrid model for Ethereum
Buterin argued that a future Ethereum will be called a blockchain largely for historical reasons. In practice, he wrote, it would combine Bitcoin’s founding ideas with cryptographic tools drawn from 50 years of academic research, including techniques that were not mature when Bitcoin launched in 2009.
In the essay, he annotated the Bitcoin whitepaper section by section. Nodes that currently download and re-execute every block would instead sample data through PeerDAS and verify a SNARK, a short proof showing that a computation was carried out correctly.
He also said block construction would move away from a model where a single miner or proposer builds the block, toward one split across multiple parties.
Performance targets in the post
A comparison table in the essay places Ethereum in 2030 at roughly 4 to 8 seconds per slot and 8 to 32 seconds to finality. As a reference point, the post compares that with about 17 seconds per block and around 200 seconds for 12 confirmations in 2015.
Decentralization as a source of speed
Buterin wrote that the biggest shift is that decentralization is moving from being purely a burden accepted for safety and robustness to, in some limited cases, a source of speed.
A distributed network can store data and run computation in parallel, he said. Early Ethereum designs tried to divide work this way, but they could not verify that each piece had been executed correctly. With modern cryptography, he wrote, that problem is now solved, and the overhead from that solution is falling month by month.
He also said developers will face different incentives. Computation packed into one serial transaction will cost more than work split into parallel pieces.
Buterin described efficient and safe zero-knowledge proofs as difficult, though already seeing heavy optimization with AI tools. Handling very large amounts of state, and allowing many processes to read and write that state at the same time, is likely the harder problem, he wrote.
How the essay fits Ethereum’s roadmap
The essay builds on Lean Ethereum, the rebuild Buterin has previously said would take three to four years and would rival the Merge in scope.
It also follows the strawmap roadmap introduced in February by Ethereum Foundation researcher Justin Drake.
Looking further ahead, Buterin said practical program obfuscation could bring another shift, though he added that the essay’s main conclusions would apply long before that becomes available.