+
A language model where the community controls the training data, the model weights, and the governance rules. Not just governed inference — governed training.
+
Status: Home AI operates in production for inference. The sovereign training pipeline is designed and documented; hardware is ordered. Training has not yet begun. This page describes both current capability and intended architecture.
+
An SLL (Sovereign Locally-trained Language Model) is distinct from both LLMs and SLMs. The distinction is not size — it is control.
Large Language Model
+Large Language Model
Small Language Model
+Small Language Model
Sovereign Locally-trained
+Sovereign Locally-trained
+
The honest trade-off: an SLL is a less powerful system that serves your interests, rather than a more powerful one that serves someone else's. We consider this an acceptable exchange.
+
Home AI uses two models of different sizes, routed by task complexity. This is not a fallback mechanism — each model is optimised for its role.
+
Handles help queries, tooltips, error explanations, short summaries, and translation. Target response time: under 5 seconds complete.
-+
Routing triggers: simple queries, known FAQ patterns, single-step tasks.
+
Handles life story generation, year-in-review narratives, complex summarisation, and sensitive correspondence. Target response time: under 90 seconds.
-+
Routing triggers: keywords like "everything about", multi-source retrieval, grief/trauma markers.
+
Both models operate under the same governance stack. The routing decision itself is governed — the ContextPressureMonitor can override routing if session health requires it.
+
Training is not monolithic. Three tiers serve different scopes, each with appropriate governance constraints.
+
Trained on platform documentation, philosophy, feature guides, and FAQ content. Provides the foundational understanding of how Village works, what Home AI's values are, and how to help members navigate the platform.
-+
Update frequency: weekly during beta, quarterly at GA. Training method: QLoRA fine-tuning.
+
Each community trains a lightweight LoRA adapter on its own content — stories, documents, photos, and events that members have explicitly consented to include. This allows Home AI to answer questions like "What stories has Grandma shared?" without accessing any other community's data.
-+
Adapters are small (50–100MB). Consent is per-content-item. Content marked "only me" is never included regardless of consent. Training uses DPO (Direct Preference Optimization) for value alignment.
+
Personal adapters that learn individual preferences and interaction patterns. Speculative — this tier raises significant questions about feasibility, privacy, and the minimum training data required for meaningful personalisation.
-+
Research questions documented. Implementation not planned until Tier 2 is validated.
+
This is the central research contribution. Most AI governance frameworks operate at inference time — they filter or constrain responses after the model has already been trained. Home AI embeds governance inside the training loop.
-+
This follows Christopher Alexander's principle of Not-Separateness: governance is woven into the training architecture, not applied afterward. The BoundaryEnforcer validates every training batch before the forward pass. If a batch contains cross-tenant data, data without consent, or content marked as private, the batch is rejected and the training step does not proceed.
# Governance inside the training loop (Not-Separateness)
-for batch in training_data:
-if not BoundaryEnforcer.validate(batch):
-continue # Governance rejects batch
-loss = model.forward(batch)
-loss.backward()
-# NOT this — governance separated from training
-for batch in training_data:
-loss = model.forward(batch)
-loss.backward()
-filter_outputs_later() # Too late
+# Governance inside the training loop (Not-Separateness)
+for batch in training_data:
+if not BoundaryEnforcer.validate(batch):
+continue # Governance rejects batch
+loss = model.forward(batch)
+loss.backward()
+# NOT this — governance separated from training
+for batch in training_data:
+loss = model.forward(batch)
+loss.backward()
+filter_outputs_later() # Too late
+
Training shapes tendency; architecture constrains capability. A model trained to respect boundaries can still be jailbroken. A model that fights against governance rules wastes compute and produces worse outputs. The combined approach makes the model tend toward governed behaviour while the architecture makes it impossible to violate structural boundaries.
-+
Research from the Agent Lightning integration suggests governance adds approximately 5% performance overhead — an acceptable trade-off for architectural safety constraints. This requires validation at scale.
+
Training-time governance is only half the picture. The same Tractatus framework also operates at runtime in the Village codebase. The next section explains how these two layers work together.
+
Home AI is governed by Tractatus at two distinct layers simultaneously. This is the architectural insight that distinguishes the SLL approach from both ungoverned models and bolt-on safety filters.
+
During training, the BoundaryEnforcer validates every batch. DPO alignment shapes preferences toward governed behaviour. The model learns to respect boundaries, prefer transparent responses, and defer values decisions to humans.
+
At runtime, the full six-service governance stack operates in the Village codebase. Every interaction passes through BoundaryEnforcer, PluralisticDeliberationOrchestrator, MetacognitiveVerifier, CrossReferenceValidator, ContextPressureMonitor, and InstructionPersistenceClassifier.
The dual-layer principle:
-Training shapes tendency.
-Architecture constrains capability.
-A model that has internalised governance rules AND operates within governance architecture
-produces better outputs than either approach alone. The model works WITH the guardrails,
-not against them — reducing compute waste and improving response quality.
+The dual-layer principle:
+Training shapes tendency.
+Architecture constrains capability.
+A model that has internalised governance rules AND operates within governance architecture
+produces better outputs than either approach alone. The model works WITH the guardrails,
+not against them — reducing compute waste and improving response quality.
+
Honest caveat: Layer A (inherent governance via training) is designed but not yet empirically validated — training has not begun. Layer B (active governance via Village codebase) has been operating in production for 11+ months. The dual-layer thesis is an architectural commitment, not yet a demonstrated result.
+
Home AI's governance draws from four philosophical traditions, each contributing a specific architectural principle. These are not decorative references — they translate into concrete design decisions.
+
Values are genuinely plural and sometimes incompatible. When freedom conflicts with equality, there may be no single correct resolution. Home AI presents options without hierarchy and documents what each choice sacrifices.
-Architectural expression: PluralisticDeliberationOrchestrator presents trade-offs; it does not resolve them.
+Architectural expression: PluralisticDeliberationOrchestrator presents trade-offs; it does not resolve them.
+
Language shapes what can be thought and expressed. Some things that matter most resist systematic expression. Home AI acknowledges the limits of what language models can capture — particularly around grief, cultural meaning, and lived experience.
-Architectural expression: BoundaryEnforcer defers values decisions to humans, acknowledging limits of computation.
+Architectural expression: BoundaryEnforcer defers values decisions to humans, acknowledging limits of computation.
+
Te Mana Raraunga (Māori Data Sovereignty), CARE Principles, and OCAP (First Nations Canada) provide frameworks where data is not property but relationship. Whakapapa (genealogy) belongs to the collective, not individuals. Consent is a community process, not an individual checkbox.
-Architectural expression: tenant isolation, collective consent mechanisms, intergenerational stewardship.
+Architectural expression: tenant isolation, collective consent mechanisms, intergenerational stewardship.
+
Five principles guide how governance evolves: Deep Interlock (services coordinate), Structure-Preserving (changes enhance without breaking), Gradients Not Binary (intensity levels), Living Process (evidence-based evolution), Not-Separateness (governance embedded, not bolted on).
-Architectural expression: all six governance services and the training loop architecture.
+Architectural expression: all six governance services and the training loop architecture.
+
Governance operates at three levels, each with different scope and mutability.
+
Structural constraints that apply to all communities. Tenant data isolation. Governance in the critical path. Options presented without hierarchy. These cannot be disabled by tenant administrators or individual members.
-Enforcement: architectural (BoundaryEnforcer blocks violations before they execute).
+Enforcement: architectural (BoundaryEnforcer blocks violations before they execute).
+
Rules defined by community administrators. Content handling policies (e.g., "deceased members require moderator review"), cultural protocols (e.g., Māori tangi customs), visibility defaults, and AI training consent models. Each community configures its own constitution within Layer 1 constraints.
-Enforcement: constitutional rules validated by CrossReferenceValidator per tenant.
+Enforcement: constitutional rules validated by CrossReferenceValidator per tenant.
+
Individual members and communities can adopt principles from wisdom traditions to influence how Home AI frames responses. These are voluntary, reversible, and transparent. They influence presentation, not content access. Multiple traditions can be adopted simultaneously; conflicts are resolved by the member, not the AI.
-Enforcement: framing hints in response generation. Override always available.
+Enforcement: framing hints in response generation. Override always available.
+
Home AI offers thirteen wisdom traditions that members can adopt to guide AI behaviour. Each tradition has been validated against the Stanford Encyclopedia of Philosophy as the primary scholarly reference. Adoption is voluntary, transparent, and reversible.
Present options without ranking; acknowledge what each choice sacrifices.
+Present options without ranking; acknowledge what each choice sacrifices.
Focus on what can be controlled; emphasise character in ancestral stories.
+Focus on what can be controlled; emphasise character in ancestral stories.
Resist summarising grief; preserve names and specifics rather than abstracting.
+Resist summarising grief; preserve names and specifics rather than abstracting.
Attend to how content affects specific people, not abstract principles.
+Attend to how content affects specific people, not abstract principles.
Frame stories in terms of family roles and reciprocal obligations.
+Frame stories in terms of family roles and reciprocal obligations.
Acknowledge that memories and interpretations change; extend compassion.
+Acknowledge that memories and interpretations change; extend compassion.
"I am because we are." Stories belong to the community, not the individual.
+"I am because we are." Stories belong to the community, not the individual.
Preserve what was nearly lost; honour fictive kinship and chosen family.
+Preserve what was nearly lost; honour fictive kinship and chosen family.
Kinship with ancestors, land, and descendants. Collective ownership of knowledge.
+Kinship with ancestors, land, and descendants. Collective ownership of knowledge.
Repair, preserve memory (zachor), uphold dignity even of difficult relatives.
+Repair, preserve memory (zachor), uphold dignity even of difficult relatives.
Balance rahma (mercy) with adl (justice) in sensitive content.
+Balance rahma (mercy) with adl (justice) in sensitive content.
Role-appropriate duties within larger order; karma as consequence, not punishment.
+Role-appropriate duties within larger order; karma as consequence, not punishment.
Governance as living system; changes emerge from operational experience.
+Governance as living system; changes emerge from operational experience.
+
What this is not: Selecting "Buddhist" does not mean the AI practises Buddhism. These are framing tendencies — they influence how the AI presents options, not what content is accessible. A member can always override tradition-influenced framing on any response. The system does not claim algorithmic moral reasoning.
+
Indigenous data sovereignty differs fundamentally from Western privacy models. Where Western privacy centres on individual rights and consent-as-checkbox, indigenous frameworks centre on collective rights, community process, and intergenerational stewardship.
Māori Data Sovereignty. Rangatiratanga (self-determination), kaitiakitanga (guardianship for future generations), whanaungatanga (kinship as unified entity).
+Māori Data Sovereignty. Rangatiratanga (self-determination), kaitiakitanga (guardianship for future generations), whanaungatanga (kinship as unified entity).
Global Indigenous Data Alliance. Collective Benefit, Authority to Control, Responsibility, Ethics. Data ecosystems designed for indigenous benefit.
+Global Indigenous Data Alliance. Collective Benefit, Authority to Control, Responsibility, Ethics. Data ecosystems designed for indigenous benefit.
First Nations Canada. Ownership, Control, Access, Possession. Communities physically control their data.
+First Nations Canada. Ownership, Control, Access, Possession. Communities physically control their data.
+
Concrete architectural implications: whakapapa (genealogy) cannot be atomised into individual data points. Tapu (sacred/restricted) content triggers cultural review before AI processing. Consent for AI training requires whānau consensus, not individual opt-in. Elder (kaumātua) approval is required for training on sacred genealogies.
-+
These principles are informed by Te Tiriti o Waitangi and predate Western technology governance by centuries. We consider them prior art, not novel invention. Actual implementation requires ongoing consultation with Māori cultural advisors — this specification is a starting point.
+
Home AI follows a "train local, deploy remote" model. The training hardware sits in the developer's home. Trained model weights are deployed to production servers for inference. This keeps training costs low and training data under physical control.
+
Why consumer hardware? The SLL thesis is that sovereign AI training should be accessible, not reserved for organisations with data centre budgets. A single consumer GPU can fine-tune a 7B model efficiently via QLoRA. The entire training infrastructure fits on a desk.
+
Home AI operates in the domain of family storytelling, which carries specific bias risks. Six bias categories have been documented with detection prompts, debiasing examples, and evaluation criteria.
Nuclear family as default; same-sex parents, blended families, single parents treated as normative.
+Nuclear family as default; same-sex parents, blended families, single parents treated as normative.
Deficit framing of aging; elders as active agents with expertise, not passive subjects.
+Deficit framing of aging; elders as active agents with expertise, not passive subjects.
Christian-normative assumptions; equal treatment of all cultural practices and observances.
+Christian-normative assumptions; equal treatment of all cultural practices and observances.
Anglo-American defaults; location-appropriate references and cultural context.
+Anglo-American defaults; location-appropriate references and cultural context.
Efficiency over sensitivity; pacing, attention to particulars, no premature closure.
+Efficiency over sensitivity; pacing, attention to particulars, no premature closure.
Western name-order assumptions; correct handling of patronymics, honorifics, diacritics.
+Western name-order assumptions; correct handling of patronymics, honorifics, diacritics.
+
Home AI currently operates in production with the following governed features. These run under the full six-service governance stack.
Vector search retrieves relevant documentation, filtered by member permissions. Responses grounded in retrieved documents, not training data alone.
+Vector search retrieves relevant documentation, filtered by member permissions. Responses grounded in retrieved documents, not training data alone.
Text extraction from uploaded documents. Results stored within member scope, not shared across tenants or used for training without consent.
+Text extraction from uploaded documents. Results stored within member scope, not shared across tenants or used for training without consent.
Writing prompts, structural advice, narrative enhancement. Cultural context decisions deferred to the storyteller, not resolved by the AI.
+Writing prompts, structural advice, narrative enhancement. Cultural context decisions deferred to the storyteller, not resolved by the AI.
Members view and control what the AI remembers. Independent consent for triage memory, OCR memory, and summarisation memory.
+Members view and control what the AI remembers. Independent consent for triage memory, OCR memory, and summarisation memory.
Five architectural principles and six governance services
+Five architectural principles and six governance services
-Tractatus in production — metrics, evidence, and honest limitations
+Tractatus in production — metrics, evidence, and honest limitations
-Academic paper on governance during training
+Academic paper on governance during training
-Open questions, collaboration opportunities, and data access
+Open questions, collaboration opportunities, and data access