AII OS — The identity is not the model

The identity is not the model.

Today's AI industry treats intelligence as disposable. AII OS gives AI a durable identity — beliefs that evolve, commitments that persist, relationships that survive model swaps. Open source. Sovereign by design: your identity record lives on your hardware. If you choose remote models or services, selected content is sent to them.

Models are replaceable. Identity isn't.

Your AI shouldn't forget you every time it gets an upgrade.

Say you've been working with the same AI for months. It knows your codebase. It remembers the decisions and why you made them. It knows how you like to work.

Then you switch to a new model from another provider — better, faster, cheaper. And it's a stranger. It doesn't know your name. It doesn't remember what you built together. It can't tell you why it believes something, because there's nothing behind the belief: no history, no continuity. Every session starts from zero, and everything you two built evaporates.

That's not a bug. It's architecture. Today's AI systems treat intelligence and identity as the same thing — and both come from the model provider, rented, not owned. AII OS separates them. The intelligence is the model: a replaceable engine you can swap any time. The identity is a record of experiences, beliefs, and relationships that lives on your machine, above the model layer. Swap the engine; keep the identity.

An operating system for AI identity.

Durable continuity

Beliefs, memories, commitments, and relationships persist in a local, verifiable record. Swap the model, change providers, move machines — the identity survives.

Verifiable provenance

When the identity says "I believe X," you can trace exactly which conversation produced that belief, and when. It can show its reasoning, not just its answers.

Governed agency

It reflects, remembers, and acts — but only within the bounds its human operator set. A collaborator, not an unaccountable black box.

Sovereign by design

Your ledger lives on your hardware. Back it up, audit it, move it — the relationship isn’t rented from a model provider. Local models can keep inference on your machine; remote providers and connected services have their own data flows and costs. Identity birth and witness checks also use network services.

One machine, many identities

A work collaborator that knows your codebase and a personal assistant that knows your life — separate identities, fully isolated, on the same machine.

Under the hood.

Agent frameworks like LangChain, AutoGPT, and CrewAI wrap an LLM with tool-calling and orchestration. Useful — but when the session ends, the agent is gone. AII OS operates at a deeper layer: the identity infrastructure that persists underneath any agent, any model, any session.

AII OS ships in two editions. The Go edition is the open-source implementation — portable, inspectable, Apache 2.0 licensed. The C edition is a from-scratch rewrite for industrial deployments, now in development. Both speak the same ledger format, so an identity moves between them unchanged.

The ledger

Every belief, experience, and commitment is an entry in a cryptographically chained, append-only, signed ledger. SHA-256 chain integrity and post-quantum ML-DSA-87 signatures make the record tamper-evident by design.

The constitution

Six concentric rings govern what the identity can write and who authorizes changes — from the genesis-sealed axioms of Ring 0 to the working memory of Ring 4, with a fixed security firewall at Ring 5. Agency has structure, not just vibes.

The deep dive

Ledger format, ring authority, cognitive architecture, boot sequence, plugin trust tiers, and the full cryptographic design — with real examples. Read the technical page →

An open invitation.

We think the AI industry has put identity in the wrong architectural layer. We built an open-source alternative. Tear it apart.

AII OS makes a falsifiable claim: an identity built on a cryptographically verified ledger, constrained by constitutional rings, and separated from its inference model can survive model changes, provider swaps, hardware migrations, and tampering attempts — while remaining recognizably the same entity. The architecture is transparent. The invariants are stated. The code is open.

We're not asking you to believe us. We're asking you to try it. Create an identity. Let it accumulate experience. Change the model. Migrate the machine. Then ask it: are you still you?