All editions · Family Edition

Your Community, Your AI — CC BY 4.0

What's Actually Running in Village Today

Early Days

This article is about what exists today — not what we plan to build, not what we hope to achieve, but what is running right now in production. Where something is planned but not yet live, we say so plainly. (Any unfamiliar term is defined in plain language in the glossary.)

Village AI has been running in production since 2025. It is a young system: some parts work well, some parts are still rough, and this article marks both. A family that adopts a platform on clear information makes a steadier partner than one that adopts on marketing.

What Village AI Can Do for Your Family Today

Answer questions about your family's content. When a family member asks "When was Grandma and Grandad's wedding anniversary?" or "What did Uncle David write about his time in the navy?", Village AI searches your family's actual records — stories, photos, event descriptions, shared documents — and provides an answer grounded in that content. It does not guess or fill in from general knowledge. If it cannot find the answer in your records, it says so.

Help with drafting. Village AI can help draft family newsletters, event invitations, and messages. Because it has been trained on your family's previous content, its drafts reflect your family's tone and style — not a generic template. A family coordinator reviews and edits every draft before it reaches the family.

Summarise long documents. A lengthy collection of family stories or a series of updates from different branches of the family can be summarised into key points. This is useful for family members who want to stay connected but do not have time to read everything.

Translate between languages. Village supports five languages — English, German, French, Dutch, and Te Reo Māori. The AI assists with translation of family content, though human review is recommended for important communications.

Triage family feedback. When a family member submits feedback through the platform — a question, a suggestion, a report of something not working — the AI classifies it, investigates where possible, and notifies the member when it has been addressed. This is designed to happen without the family coordinator having to sort every piece of feedback by hand.

Where Village Acts, Not Just Answers

Articles 1 and 2 made a distinction worth holding onto: a chatbot answers, an agent acts. So it is fair to ask — does Village AI only answer, or does it act too?

It acts, in carefully bounded ways. The clearest example today is feedback resolution. When a family member marks an answer as unhelpful, Village does not simply log a complaint for a person to wade through later. It investigates the correct answer against your family's records, and where it can resolve a routine, low-stakes case on solid evidence, it does so on its own — improving the system's knowledge so the next family member who asks gets a better answer. This is genuine agentic behaviour: the AI takes a multi-step action, not just produces a sentence.

But notice the boundaries around it, because they are the whole point. The agent acts only inside your family's data. It acts only on routine, reversible matters. And the moment it detects a systemic problem — a pattern of related failures that suggests something deeper is wrong — it stops acting and routes it to a person, because that is no longer a routine fix but a judgement call. The design target is that the great majority of ordinary feedback is handled automatically, while anything consequential lands on a person's desk.

This is the practical shape of the principle from Article 3: an agent you control acts where action is safe and reversible, and steps back where it is not. That is the opposite of an outside agent that acts on everything with fewer chances for you to step in.

What the AI Does Not Do

It does not make decisions for your family. When a question involves values, privacy, or judgement, the AI stops and routes it to a person. Your family coordinator, your family elder, the family as a whole — the people your family trusts with these decisions.

It does not take consequential or irreversible actions on its own. As the previous section described, where Village AI acts, it acts only on routine, reversible matters inside your family's data. It does not send messages in your name, commit your family to anything, or make changes it cannot walk back, without a person in the loop. The keys stay with your family.

It does not access content it was not given. Private content stays private. Content from other families stays with those families. The AI cannot reach across boundaries, because those boundaries are structural, not policy-based.

It does not operate without oversight. Every AI response passes through Guardian Agents — the independent verification layers described in the previous article. No response reaches a family member without being checked against your family's actual records.

It does not pretend to know things it does not know. When the AI's confidence is low, it says so. Every response carries a confidence indicator. Family members can see at a glance whether the AI is drawing on solid records or venturing into less certain territory.

How Bias Is Addressed: The Vocabulary System

One of the subtlest forms of bias in AI is linguistic. When a system trained on corporate data calls your family members "users" and your family stories "posts," it is imposing a worldview — one where families are consumer platforms and communication is content marketing.

Village addresses this through a vocabulary system that adapts the entire platform to your family.

When you set up a Village for your family, the system does not show you generic labels. It shows you the language of family life:

• Family members, not "users" or "subscribers"
• Family stories, not "posts" or "updates"
• Family records, not "admin settings"
• Family news, not "content"
• The family, not "the community workspace"

This is not cosmetic. The vocabulary shapes how the AI understands and responds to your family. When the AI has been trained with the term "family member" rather than "user," it processes questions and generates responses within a family frame of reference. It understands that "How do I share Grandma's recipe with everyone?" is a different question from "How do I distribute content to users?" — even though a generic AI system would treat them identically.

Each community type has its own vocabulary. A sports club sees "club members" and "season fixtures." A parish sees "parishioners" and "parish bulletins." The platform is the same, but the language — and therefore the AI's understanding — is specific to your family.

How the AI Learns and Improves

Village AI is not static. It improves over time through three mechanisms:

Scheduled retraining. The AI is retrained on your family's latest content on a regular schedule. New stories, new photos, new event descriptions — they enter the AI's knowledge base so it stays current with your family's life.

Family coordinator feedback. When a coordinator flags an AI response as inaccurate or unhelpful, that correction feeds back into the system. Over time, the AI learns what works for your family and what does not. This is not generic improvement — it is improvement specific to your family.

Guardian-agent learning. An adaptive layer adjusts verification thresholds based on patterns of accuracy and error. If the AI consistently gets a certain type of question right, the guardian eases verification intensity for that type. If it consistently struggles with another type, the guardian tightens scrutiny. The system becomes more efficient without becoming less careful.

What Is Still a Work in Progress

A higher-capability tier. Village runs a focused open model, fine-tuned per community type, so a family, a parish, and a club each get an AI tuned to their context — for a family, one shaped around family life across the generations. The architecture reserves a more capable tier for harder questions; it is defined but not yet in service, so today every question is answered by the community-tuned model.

Individual personalisation — where the AI learns individual family member preferences — is planned but not yet built. For now, the AI knows your family as a whole, not your individual family members as individuals (unless they interact with it directly).

The coordinator accreditation path — structured training for family members who take on the coordinator role — is designed but being rolled out progressively. Founding families have direct access to the founder for support.

We mention these plainly because you should know what you are adopting. This is a young platform, built by a small team and used by a small number of families. It is functional, it is improving, and it is clear about where it stands.

What This Means for Your Family

If your family is considering Village, here is what you are choosing:

Village is a platform where AI knows your family's actual content — your stories, your photos, your events — not the internet's idea of what a family might be. Every AI response is mathematically verified against your records by independent watchers. The vocabulary reflects your family: family members, not users; family stories, not content.

Your data stays within your family's boundary, is not used to train external AI systems, and can be exported or deleted at any time. The system is transparent about its limitations, improves from your coordinators' corrections, and stops to ask a person when a question requires judgement rather than information.

You would also be joining as a founding family — one of a small number of families, parishes, clubs, and organisations shaping the platform in its early life.

If that interests you, the current status of the founding-community programme is on the founding-community page.

Want to use AI tools like these well, and safely? Our free courses — Working with Claude and Agents at Work — teach the practical skills, from getting trustworthy answers to deciding what to hand an agent. For the full technical architecture behind Village AI, see Village AI — Agentic Governance.