The Sovereignty Foundation™
The Sovereignty Stack
One pipeline, movable boundaries — the architecture at a glance, before the whitepapers
LOGOS™ is the identity system — the entire pipeline that continually validates identity through gated execution. It ingests signals through AXIS, enforces role and policy through RITA™/SAGE™ where the authorization decision is minted, hands the decision to KNOX™ — the notary and cryptographic authority, with temporal validation — and enforces the execution boundary through AEGIS™. What travels through the pipeline is the WISP™: the cryptographically-bound identity object the system produces and continually re-measures. It is not a credential — and that is the filed invariant: no separable, transferable, independently-usable artifact exists at any interface, and no execution path bypasses identity-object presence. Beneath it all runs the SOV substrate: Canon, Lexicon, Pulse Mesh, Musashi/Kronos, and Hermes.
One machine, air-gapped: every stage and the substrate inside a single physical boundary — a filed embodiment, not a degraded mode. Artifacts can cross by physical media.
Why the boundaries move
Every hand-off in the pipeline is a fixed-format, self-proving artifact — verified where it arrives, never trusted for where it came from. That single discipline makes the trust boundaries free variables: the same system runs entirely inside one process on one air-gapped machine, or with signal processing on one boundary and RITA™/SAGE™ on another, or everything local with only KNOX™ across the line and execution bound back — or fully distributed across processes, containers, hosts, clouds, and hardware-attested enclaves. The topology is a deployment decision, not an architectural one, and this dynamic composition is itself part of what is patented: air-gap operability and hardware-attested execution are filed embodiments, and the separation of governance, notary, and execution zones is a claimed system family.
Go deeper
The fourteen functional surfaces (detail)
- Identity — WISP composite, modality vectors, coherence evaluation
- Semantic memory — Canon hippocampus + neocortex, HNSW retrieval, decay
- Cryptographic authorization — Knox warrants, VDFs, RSW time-locks, memory-hard PoW
- Verification — Aegis five-stage gauntlet, integrity sealing, hash-chain attestation
- Wire transport — Pulse Mesh deterministic 128-byte format, seqlock concurrency
- Time — Kronos Fourier-encoded 8D temporal embeddings, half-life decay
- Governance — deterministic decision engine, hash-chained decisions, fail-closed semantics
- Observability — passive multi-level telemetry, provenance reconstruction, Owl ledger
- Substrate shape — Lexicon shape catalog, deterministic descriptors, BLAKE3 Merkle tree
- Standards binding — SIG-ID bridge primitive, code-to-spec-to-patent linkage
- Multi-zone boundary — three-zone physically-separated security model
- Topology — node identity, ring registry, mesh routing
- Spell & scroll — composable execution units, character-creator ontology
- Memory graph — IP and substrate provenance, JSON-LD knowledge representation
Approximately three hundred distinct primitives are documented across these surfaces, each with a stable SIG-* identifier, an implementation pointer, and a formal description. Built and tested, not theoretical.
Why this exists
Most AI infrastructure published today has the same shape: a probabilistic model, an API surface, a set of guardrails layered on top, and a set of policies asserting how the system should behave. The Foundation’s view is that this shape cannot provide the deterministic authorization, provenance and replay properties this architecture defines as computational governance — because the controls are asserted at inference time and are themselves probabilistic.
The stack proposes an alternative shape: governance enforced structurally, before runtime, through deterministic primitives whose behavior is cryptographically auditable. The pipeline above exists to demonstrate that the proposed governance posture is mechanically possible, not aspirational.
Governance must be enforced through architecture rather than asserted through policy.