LayerZero Verified Most RISC-V Bytecode Instructions
The team successfully proved 60 of 67 RISC-V instructions for zero-knowledge proofs in a 2.5-month effort.
Updated on Sept. 23, 2026 in Quantum Computing

Live Poll
Do you trust AI-generated proofs to ensure software security as much as human-written ones?
LayerZero Research has verified 60 of 67 RISC-V instructions for its Jolt bytecode expansion using the Lean theorem prover. The research-stage effort ensures the mathematical validity of zero-knowledge proofs built on the RISC-V architecture.
Why it matters
Formal verification is critical for ensuring that zero-knowledge proofs—cryptographic methods used to prove data validity without revealing it—function correctly at the instruction level. Correcting these foundations minimizes vulnerabilities in high-assurance computation environments.
The team achieved a 60-instruction validation milestone using the Lean theorem prover checked against the LeanRV64D reference model. This effort required 2.5 months to complete, successfully covering the majority of the total 67 expandable RISC-V instructions.
The players
LayerZero Research
A research organization focused on cryptographic protocols and formal verification methods for zero-knowledge proof architectures.
The details
Engineers used AI tools including Claude and Codex to accelerate the generation of mathematical proofs for the Jolt bytecode expansion. Human experts authored the critical definitions and initial proof templates, while the Lean theorem prover—a software environment for formalizing mathematical proofs—ensured consistency against the LeanRV64D architecture model. Seven instructions remain unverified due to complex edge cases that require additional refinement.
Timeline
July 2026: The verification effort began.
September 2026: Formal verification results were announced.
The Tech Race
This verification effort aligns with broader industry trends to secure complex cryptographic instruction sets through machine-assisted formal proofs. By validating the Jolt bytecode, LayerZero Research aims to establish a high-assurance foundation for future zero-knowledge implementations.
This development currently impacts developers and researchers building zero-knowledge proof systems on RISC-V infrastructure. There is no public product or consumer-facing tool available as the remaining seven instructions currently lack formal verification.
The takeaway
Formal verification marks a shift toward mathematically ensuring the security of underlying bytecode for cryptographic proofs. Researchers should monitor future updates from LayerZero regarding the status of the final seven unverified instructions.
Further reading
For more on the intersection of cryptography and hardware, explore our Quantum Computing section.
Live Poll
Do you trust AI-generated proofs to ensure software security as much as human-written ones?






