Ask Church routes anyone in real distress to 988 and your own care team before any AI runs, keeps everyday questions anonymous, and only identifies someone when they choose to share their info.
A chatbot on a church website meets people at their most vulnerable. Two things have to be right before anything else:
This is the question that should decide whether you trust a church AI at all. With Ask Church, a message signaling acute distress — in English or Spanish — triggers a hard-coded gate that runs *first*, before any AI generates a word. It routes the person to:
The assistant never tries to be the counselor. It gets a hurting person to a real one. This is wired into the code to run before the AI, not handled by asking the model to be careful.
Everyday questions are anonymous — no names, no accounts, no sign-in. Someone is identified only when *they choose* to share their information to receive prayer follow-up, to serve, or to plan a visit. When they do, that information goes to your church's team — not to us. We don't build a profile of your congregation, and we don't train a model on your members' conversations.
A general chatbot bolted onto a church site has neither of these. It will try to answer a crisis message like any other prompt, and it has no concept of routing a hurting person to a human. Crisis safety and privacy aren't features you turn on later — they're decisions made in the architecture, before the first answer.
Yes. The gate detects acute distress in English and Spanish and routes before any AI runs.
Everyday questions are anonymous and used only to improve the church's own assistant. We don't attach identities to them, and identity is only ever shared by the person themselves, to your team.
Yes — the church gets an anonymized view of what people asked (content yes, identity no), so you can see the questions your community has without knowing who asked.
Updated 2026-06-19 · Ask Church by Vision Genesis · Knoxville, TN
See it answer — try a live demo →