In an announcement that rekindles the perennial debate over how far Bitcoin can—or should—stretch beyond its original remit, Input Output (IO), the research and engineering company behind Cardano, has unveiled a compilation pipeline that executes Cardano smart contracts on Bitcoin without touching a line of Bitcoin’s consensus code.
“A novel compilation pipeline being developed at Input | Output takes a Cardano smart contract, written using a modern language like Python, TypeScript, or Scala, and runs it on Bitcoin without modifying Bitcoin itself,” the company wrote on X late Tuesday night, attaching a six‑post technical thread that spelled out the process in detail.
Bitcoin Gets Cardano Smart Contracts
The thread describes a relay of representations that begins with high‑level source code and ends inside Bitcoin Script. First, Cardano’s tool‑chain lowers the contract into Untyped Plutus Core (UPLC), a stripped‑down functional bytecode. UPLC is then serialized and handed to a miniature interpreter known as the Control–Environment–Continuation (CEK) machine. Instead of asking Bitcoin to interpret UPLC directly, IO compiles the CEK machine itself into RISC‑V, “a real‑world, simple CPU instruction set.” The resulting RISC‑V program becomes the payload that Bitcoin ultimately verifies, with the serialized UPLC passed to it at run‑time.
A companion blog post signed by software engineer Riley Kilgore frames the design in a single sentence: “By combining a serialized smart contract format (Untyped Plutus Core – UPLC), a clever interpreter architecture (CEK machine), and a widely supported open‑source reduced instruction set architecture (RISC‑V), IO is turning that idea into reality.” Because the RISC‑V code is deterministic and its execution trace can be proven—or disputed—inside Bitcoin Script, no soft‑fork is required. Bitcoin remains unchanged; the expressiveness is off‑loaded to an auxiliary virtual CPU whose behaviour can be challenged step by step.
The security anchor for those challenges is BitVMX, a descendant of Robin Linus’s 2023 BitVM proposal for “disputable computation” on Bitcoin. IO’s newsroom statement calls BitVMX an “innovative, trust‑minimized protocol” that “operates on a 1‑out‑of‑n honest‑operator model” and lets developers “pay transaction fees in BTC directly from their preferred wallet.” BitVMX inherits the optimistic‑roll‑up style of BitVM—off‑chain execution, on‑chain fraud proofs—but replaces BitVM’s bespoke circuits with a full RISC‑V core, allowing unmodified binaries compiled by standard tool‑chains.
The promise is to bring Cardano’s eUTxO‑based smart‑contract ecosystem—Marlowe, Aiken, and the rest—into Bitcoin’s vast liquidity pool. In practical terms that could mean lending, swaps, and even NFT‑backed loans secured by BTC collateral, all without giving custodial control to a sidechain.
For Bitcoin purists the move is both familiar and contentious. Taproot in 2021 expanded Script with Schnorr signatures and key‑path spends, giving the network its first taste of complex smart‑contract logic, while BitVM in 2023 showed that arbitrary computation could be verified rather than executed by Bitcoin nodes. IO’s pipeline pushes those ideas further, but it does so in a way that keeps the base layer unaltered—a design criterion that BitVM’s authors called non‑negotiable for Bitcoin’s social contract.
The compilation flow is still in active development; IO has not committed to a main‑net launch date, nor has it published gas‑cost benchmarks for typical contracts. What it has done is demonstrate a theoretically complete bridge between two philosophically divergent blockchains. If the engineering hurdles—transaction sizes, dispute latency, and user‑experience frictions—can be contained, Cardano’s software stack may soon become a first‑class citizen on Bitcoin.
At press time, ADA traded at $0.87.
