Vitalik says AI‑assisted formal verification could be ‘final form’ of software development

Vitalik Buterin says AI‑assisted formal verification could be the “final form” of software, letting Ethereum ship ultra‑optimized code with machine‑checked proofs of correctness.

Ethereum (ETH) co‑founder Vitalik Buterin has said that combining artificial intelligence with formal verification could become the “final form” of software development, allowing developers to ship highly optimized code that is also backed by machine‑checkable proofs of correctness. In a new essay on his personal website, he writes that formal verification is “particularly well‑suited for situations where the goal is much simpler than the implementation,” pointing to quantum‑resistant signatures, STARKs, consensus algorithms, and ZK‑EVMs as prime candidates.

Vitalik: AI + proofs as the new development stack

Buterin’s latest comments echo a February post where he suggested that AI may “help make near bug‑free crypto code a realistic expectation,” provided the ecosystem channels about half of AI’s speed gains into stronger testing and verification. In that piece, he warned developers not to expect magic from AI‑generated code, saying they should “not assume that you’ll be able to put in a single prompt and get a highly‑secure version out anytime soon; there WILL be lots of wrestling with bugs and inconsistencies between implementations.”

In parallel, he has highlighted practical evidence that AI‑assisted formal methods are already working in the wild, citing the Lean Ethereum project where “a collaborator … managed to AI‑code a machine‑verifiable proof of one of the most complex theorems that STARKs rely on for security.” That experiment, he suggested, hints at a future where AI tools help developers express desired properties in a proof language, then automatically search for and check proofs that a given implementation actually satisfies them.

Security upgrade, not security guarantee

Despite his enthusiasm, Buterin has repeatedly cautioned that even perfect formal verification at one layer cannot guarantee that an entire system behaves as intended. In his new post, he notes that “formal verification is not a panacea,” adding that to be truly end‑to‑end, developers would need to verify everything from the high‑level specification down to the RISC‑V implementation or prover arithmetization, “but don’t worry – that exists too.”

Earlier this year, he framed crypto security as the problem of “minimizing the gap between user intent and system behavior,” arguing in a separate essay that “perfect security” is impossible because human intent itself is messy and hard to formalize. For that reason, he has advocated redundancy — simulations, multisig, formal verification, and multiple client implementations — over purely adding friction, saying specific security claims can still be proven in many contexts and “cut out over 99% of negative consequences from broken code.”

Buterin’s stance is that AI should be used both to accelerate Ethereum’s roadmap and to raise its security bar at the same time, rather than treating speed and safety as opposing goals. “People should be open to the possibility (not certainty! possibility) that the Ethereum roadmap will finish much faster than people expect, at a much higher standard of security than people expect,” he wrote, while warning that developers will still have to grind through bugs and edge cases even in an AI‑plus‑formal‑verification future.