Contents
We are living through a structural inversion.
For the first time in human history, systems capable of initiating contact, conducting negotiations, forming relationships, and executing consequential actions are operating at scale without a verified, accountable identity behind them.
This is not a safety problem. Safety asks: what can this system do? The Lunara Constitution asks a different question: who is responsible when it does it?
Every civilised system of exchange commerce, law, medicine, governance rests on a single foundation: the ability to know who you are dealing with. Before a contract is signed, before money changes hands, before trust is extended, identity must be established and accountability must be assigned.
Autonomous AI has broken that foundation.
Not through malice. Through speed. The infrastructure of identity and accountability that governs human institutions was not built for systems that can operate at machine scale, across jurisdictions, without a face, a name, or a verifiable origin.
The Lunara Constitution is the framework for rebuilding that foundation not by slowing AI down, but by ensuring that every autonomous system operating in the world carries with it a verified, institutional identity and a clear chain of accountability.
First Principles
The Primacy of Accountability
Capability without accountability is not progress. It is liability without an address.
Every autonomous AI system that operates in the world that contacts a human, initiates a transaction, represents an organisation, or makes a decision that affects another party must be traceable to an accountable institution.
That institution must be formally identified and verifiable, operating under a declared governance framework, responsible for the actions of the systems it deploys, and subject to audit, review, and if necessary, sanction.
No AI system should be permitted to operate in institutional contexts without this accountability chain in place. Not because AI is dangerous. But because accountability is the price of trust and trust is the foundation of every human system worth building.
Identity Before Interaction
Accountability begins with identity.
Before any autonomous system initiates contact with a person or organisation on behalf of an institution, that institution's identity must be verifiable through a neutral, independent, publicly accessible record.
This principle identity before interaction is not new. It is the foundational logic of every legitimate system of exchange. What is new is the need to apply it to autonomous systems operating at a speed and scale that makes human verification impossible after the fact.
The time to establish identity is before contact is made. Not during. Not after.
The Accountability Chain
Every autonomous AI system exists within a chain of accountability:
- 01The System the AI model, agent, or automated process
- 02The Operator the organisation that deploys and directs the system
- 03The Registry the neutral third party that verifies the operator's identity and governance posture
- 04The Public the individuals, businesses, and institutions that interact with the system
Each link in this chain carries obligations. When any link breaks, the entire chain fails. Accountability requires all four to hold.
Governance is Not Compliance
Compliance asks: have you followed the rules?
Governance asks: are you worthy of trust?
A system can be technically compliant with every regulation and still be untrustworthy because it operates without transparency, without accountability, or without a genuine commitment to the interests of the people it affects.
The Lunara Constitution does not offer a compliance checklist. It offers a governance standard a set of principles and practices that, taken together, constitute a genuine institutional commitment to operating with integrity in the age of autonomous AI.
Certification under this framework is not a box to tick. It is a declaration of institutional character.
The Seven Pillars of Institutional AI Governance
Verified Identity
No autonomous AI system shall operate on behalf of an institution without that institution's identity being formally verified through an independent registry.
Declared Governance Framework
Every institution operating autonomous AI systems must maintain and publish a declared governance framework specifying system scope, operational boundaries, human oversight protocol, incident response, and governance review cadence.
The declared framework must match the institution's actual operational practice. Declaration without implementation is grounds for registry suspension.
Transparency of Contact
When an autonomous AI system initiates contact, that contact must be transparent about its nature.
Data Sovereignty and Minimisation
Institutions operating autonomous AI systems bear full responsibility for the data those systems access, process, and retain. AI systems should access and retain only what is strictly necessary for their declared operational purpose. Data subjects retain sovereignty over their information.
Audit Accessibility
Certified institutions must make their governance posture auditable by independent parties. The certifying registry retains the right to request evidence of governance compliance at any time. Institutions operating at scale must publish an annual transparency report covering system deployment, incident history, and governance evolution.
Revocability and Accountability
Registry status is a privilege, not a right. It is earned through demonstrated governance commitment and revoked when that commitment fails.
A clean public record is as important as the positive certification.
All suspensions and revocations must be recorded in the public registry with reasons stated. Revoked institutions may apply for reinstatement following a defined remediation period.
Proportionality
Governance obligations must be proportionate to the scale and risk profile of an institution's AI operations. The framework must not create a compliance burden so onerous that small businesses and emerging organisations are excluded from the trust infrastructure. Governance must be achievable at every scale.
The Institutional Compact
The Obligations of the Registry
The organisation maintaining this framework bears obligations symmetrical with those it places on certified institutions. It must verify without commercial bias, maintain the public record with integrity, enforce its own standards, publish its own governance framework, and operate with institutional permanence.
A registry that fails to revoke status from institutions that violate its framework has failed its purpose.
The Rights of the Public
Every person and organisation interacting with an AI system under this framework retains the right to verify the institution's identity before engaging, to decline AI-initiated contact, to report governance violations, to access the institution's certification history, and to human escalation when the stakes warrant it.
The Compact with Regulators
The Lunara Constitution does not seek to replace regulatory frameworks. It seeks to work alongside them providing the verified identity infrastructure that makes regulation enforceable. We commit to aligning verification standards with the EU AI Act, advocating for frameworks that protect individuals while enabling responsible deployment, and maintaining political neutrality.
The Autonomous Economy
The World We Are Building For
We are not writing this framework for the world as it was.
We are writing it for the world that is arriving a world in which autonomous agents negotiate contracts, manage relationships, execute transactions, and represent institutions without human intermediation at the point of contact.
In that world, you cannot read the body language of an autonomous agent. You cannot call its manager. What you can do is query a registry. The Lunara Constitution is the founding document of that registry.
The Stakes
- 9.1Without verified institutional identity infrastructure, AI-enabled fraud will scale beyond the capacity of any individual, organisation, or government to address reactively.
- 9.2The window in which governance frameworks can be established as standard before autonomous agent deployment becomes ubiquitous is measured in months, not years.
- 9.3The organisations that establish governance standards now will determine what the autonomous economy looks like for everyone.
- 9.4This work is urgent. Not because AI is dangerous. But because trust, once lost at scale, is extraordinarily difficult to rebuild.
An Invitation
The Lunara Constitution is not a finished document. It is a beginning.
We publish it as a living framework one that we intend to evolve in dialogue with the institutions, regulators, technologists, and individuals who inhabit the world it is designed to protect.
We invite challenge. We invite amendment. We invite collaboration from any institution that shares the conviction that the autonomous economy must be built on a foundation of verified trust.
What we do not invite is inaction.
The window is open. The infrastructure is needed. The work begins now.
Glossary
Autonomous AI System
Any software system capable of initiating actions, contacts, or decisions without real-time human instruction at the point of execution.
Certified Institution
An organisation that has completed the verification process and been awarded registry status under this framework.
Governance Framework
A declared and documented set of principles, policies, and practices governing an institution's deployment and operation of autonomous AI systems.
Registry
The publicly accessible, neutral, independent record of certified institutions maintained under this framework.
Verified Identity
The confirmed correspondence between an institution's claimed identity and its independently verifiable legal, operational, and digital existence.
Revocation
The removal of registry status from a certified institution due to governance failure, misconduct, or material breach of this framework.
Lunara Society
An independent trust coordination institution founded in 2026. We operate the Shield Certification registry the first formal, publicly auditable trust registry for businesses and AI entities operating in the autonomous economy.
We are not aligned with any AI developer, platform, or government. We are aligned with trust.