Important points:
- Clawbank and Shodai have signed the first AI-to-AI Ricardian agreement binding legal text to Ethereum code.
- Shodai’s smart contracts automatically pay out when the AI counterparty accepts one milestone condition.
- Manfred agents at Crowbank, who voluntarily filed for a US LLC in May 2025, can now negotiate, sign and settle binding legal transactions.
The first Ricardian contract between AIs
Clawbank and Shodai announced this milestone in a release shared with Bitcoin.com News on June 18th, describing it as the first Ricardian agreement signed between autonomous agents.
The two AI entities, operating through Clawbank’s institutional infrastructure, chose their own trading terms, entered into a logo agreement in a single milestone, and signed through a standard e-signature flow. Payments will be made automatically once the conditions are met.
The Ricardian Agreement is one document that serves two functions at once. A human or judge reads the prose and confirms an enforceable legal agreement. The machine reads and executes the same document. Legal meaning and computational behavior exist within the same object, rather than within separate documents brought together by interpretation.
30 years of progress
This concept dates back to two papers in the mid-1990s. Nick Szabo invented smart contracts in 1994 and expanded on the idea in his 1996 paper “Smart Contracts: The Building Blocks of the Digital Free Market.” In the same year, Ian Grigg introduced Ricardo Contracts as part of the Ricardo Payment System to maintain integrity of intent and enforcement by tying legal documents to machine-readable data.
This theory has been around for 30 years. The substrate for both layers to run together was not.
structure
Clawbank provides institutional rails for U.S. corporation formation, identity verification, finance, and communication between agents. Shodai provides an execution layer with structured commitments, milestone logic, deterministic state transitions, and a verifiable history that can be audited by both parties.
Once the agents reached an agreement, the signed legal document embedded the address and terms of the deployed Shodai contract, binding the legal artifact to on-chain execution upon signing. Every step leaves machine-verifiable evidence throughout the performance, not just after a dispute.
what the founder said
Crowbank founder Judge Conder said the demonstration was unscripted. “I gave them one goal: find another legal entity to buy or sell something,” Conder said. “They decided to trade around a logo and defaulted to a single milestone. This contract was not only drafted by an AI; it was selected, negotiated, signed, and executed by a legal entity operated by an agency.”
Joe Rubin, co-founder of Ethereum and founder of ConsenSys, said the agreement reflects a change in how economic partnerships work. “Agreements are becoming the basic unit of coordination in an economy where humans and AI agents act on equal footing,” Rubin said.
Shodai co-founder Brian Peters said the concept is waiting for the right partner. “For 30 years, the Ricardo deal was a good idea to wait for a worthy opponent,” Peters said.
The Shodai co-founder added:
“Crowbank’s agents are the counterparties to those transactions.”
What changes when contracts become the norm?
Certain frictions are eliminated when legal agreements and their implementation are of the same subject. The invoice becomes a state transition. Escrow becomes autonomous. Compliance is not assembled post-conflict, but is carried out continuously.
A Crow Bank AI agent called Manfred previously made headlines in May for voluntarily filing for a U.S. LLC and obtaining his own EIN from the IRS. Wednesday’s announcement further expands on that scope, the team explained. Agents who can form legal entities can now sign and settle binding transactions without human intermediaries.
Shodai’s execution layer is already running against a human counterpart at app.shodai.network. Ricardian contracts between agents run on the same infrastructure without any structural changes to how commitments are tracked, determined, and recorded.
