Ethereum’s long-distance protocol roadmap could move faster than many expect as AI tools improve, according to Vitalik Buterin. He pointed to a recent experiment that used agent coding to assemble an ambitious reference client that would span most of Ethereum’s planned 2030 architecture.
The comments come after developer Jiayao Qi, published as YQ via The project reaches 702,000 lines of Go, covers 65 roadmap items over 8 phases, passes 36,126 official Ethereum health tests, and can be synchronized with the mainnet through integration with go-ethereum v1.17.0. Qi said it built the client in about six days using Claude Code at a cost of about $5.75 trillion and 2.77 billion tokens.
AI can speed up the Ethereum roadmap
Buterin called the effort “a very impressive experiment,” but emphasized that prototypes built at that speed have clear limitations. “These devices, built in two weeks without an EIP, come with huge caveats,” he wrote. “There will almost certainly be a lot of critical bugs, and in some cases it may be a ‘stub’ version of something the AI didn’t even try to create a full version of. But six months ago even this was well outside the realm of possibility, and what matters is where the trend is going.”
This distinction was more important to Buterin than the raw demonstration itself. In his view, AI is not just about reducing development time. The way Ethereum engineers approach assurance may change. “Perhaps the right way to use it is to take half the speed boost from AI and half the security boost,” he said. “Generate more test cases, formally verify everything, and do more multiple implementations.”
He linked this directly to the official verification work underway for Ethereum. Referring to Lean Ethereum efforts, Buterin said one collaborator has already used AI to generate machine-verifiable evidence for the complex theorems that underpin STARK’s security. “@leanethereum’s core principle is to formally verify everything, and AI is greatly accelerating our ability to do this,” he wrote. “Besides formal verification, being able to generate a lot more test cases is also important.”
ETH2030 itself was presented as a stress test for the roadmap rather than a candidate client. Qi has repeatedly framed it as a draft rather than production software, arguing that its value lies in exposing difficult engineering problems to the public now, rather than years from now.
The roadmap implemented in the project targets a version of Ethereum with over 10,000 TPS on L1, final completion in seconds instead of 15 minutes, solo staking of 1 ETH, stateless nodes running on a $7 Raspberry Pi, and over 1 million TPS on L1 and L2. However, experiments also revealed deep coupling between upgrades, from block access lists and gas price adjustments to PeerDAS, native rollup, and fast finality.
Qi was blunt about the gap. The Pure-Go crypto implementation delays production code by roughly 10x to 100x, the consensus logic is not battle-tested on a live beacon chain, and the jump from the current roughly 5 million gases per second to the 1 billion gas per second target remains highly speculative on actual MEV and contract dependency patterns.
Buterin did not claim that AI would solve these problems. In fact, he warned us not to expect security protocols from a single prompt. “There will be a lot of wrestling with bugs and inconsistencies between implementations,” he wrote. “But even that wrestling can happen five times faster and 10 times more thoroughly.”
More important than the headline numbers are the questions now in front of Ethereum researchers and client teams. If AI can speed up both implementation and verification, your roadmap may not be just a distant architectural sketch. As Buterin said, people should at least be open to the “possibility” that Ethereum’s roadmap could be completed “much faster than people expect and with much higher security standards than people expect.”
At press time, ETH was trading at $1,956.
Featured image created with DALL.E, chart from TradingView.com
editing process for focuses on providing thoroughly researched, accurate, and unbiased content. We adhere to strict sourcing standards and each page undergoes diligent review by our team of top technology experts and seasoned editors. This process ensures the integrity, relevance, and value of the content for readers.
