The researchers argue WASM is a better long-term choice than RISC-V for Ethereum’s L1 smart contract format, or “delivery ISA.”
Offchain Labs, the core developer of Ethereum Layer 2 Arbitrum, has challenged Vitalik Buterin’s proposal to transition Ethereum’s execution layer to RISC-V, publishing a detailed technical post arguing that WebAssembly (WASM) offers superior long-term advantages.
In a Nov. 20 post on Ethereum Research, four Offchain Labs researchers contend that while RISC-V currently excels at generating ZK proofs, that doesn’t make it the right choice for how smart contracts should be delivered and stored on Ethereum.
Ethereum co-founder Buterin had floated the idea of replacing the Ethereum Virtual Machine’s bytecode with the open-source RISC-V (pronounced “risk-five”) instruction set architecture (ISA) in a post on Ethereum Magicians in April, arguing the change could slash on-chain ZK proving costs by as much as 100x in some cases.
“We support these goals but question Vitalik’s implicit assumption that one ISA can optimally serve both ZK-proving and smart contract delivery,” wrote Mario Alvarez, Matteo Campanelli, Tsahi Zidenberg and Daniel Lumi.
The team’s core argument centers on distinguishing between a “delivery ISA” (dISA) — the format for uploading contracts on-chain — and a “proving ISA” (pISA) used by ZK virtual machines. These don’t need to be the same, they argue.
Offchain Labs is already building a prototype that proves this concept: Arbitrum blocks, including WASM-based Stylus smart contracts, are ZK-proven by first compiling WASM to RISC-V and then proving the RISC-V execution.
“We can ZK-prove real-world blocks today in a blockchain that uses WASM as a dISA, by using a RISC-V-based ZK-VM as a backend,” the post states.
The team questions whether RISC-V represents the endpoint of ZK-VM evolution, noting the proving landscape is changing rapidly. Recent shifts from 32-bit to 64-bit RISC-V implementations highlight this uncertainty.
Enshrining RISC-V on L1 could lock Ethereum into a particular proving technology just as better alternatives emerge, they warn, while WASM-based ZK-VMs like Ligero’s Ligetron are already demonstrating advantages that hardware-focused ISAs may not be able to match.
Meanwhile, ZK proving costs have plummeted to around $0.025 per Ethereum block and continue falling, making it less critical to optimize exclusively for proving efficiency, the researchers argue. “Even if L1 were to require multiple ZK proofs per block, this cost would be minimal compared to the gas fees and MEV a builder could receive from a block,” they wrote.
The team highlights WASM’s structured design, which makes it easier to modify and optimize code without breaking existing contracts. WASM also executes efficiently on common hardware, while most Ethereum nodes don’t run RISC-V CPUs and would need to emulate it.
WASM’s validation capabilities ensure type safety and prevent vulnerabilities, while its mature tooling ecosystem has been battle-tested across billions of execution environments.
“We think WASM can be a sort of Internet Protocol for smart contracts, serving as an ideal intermediate layer between the diverse source-languages in which smart contracts are written and the diverse backends used to execute and prove smart contracts,” the researchers wrote.
