Our approach

Every question deserves
a faithful response.

Walking in Wisdom, unlike general AI models, always answers from a curated library, ensuring that every quote is verified before it reaches you.

An open Bible with a highlighted passage

Five steps,
every time.

This is the same path for every question, whether it is a curious first ask or a hard objection.

01 You ask

In your own words, exactly as you would ask a friend. Blunt, half-formed, or hostile is fine.

02 We search the library, not the internet

The most relevant Scripture and approved teaching is retrieved. Nothing outside the library is used.

03 The answer is written

If the library holds nothing relevant, the answer says so rather than inventing one.

04 Every quote is checked and referenced

Each excerpt is verified against its source, kept short, and credited to its author/source. A second check confirms the claim is true to what the source says. Anything that fails is rewritten or removed.

05 Insights and sources are added

Questions to take to yourself, a friend and a group, and a list of every passage, book, video or podcast used, linked so you can check them.

The difference
is measurable.

We tested where people ask: 1,000 real faith questions from Reddit (a popular online forum), 100 plain language life questions put to ChatGPT and Gemini, and pressure tests to see whether an AI will adopt a claim when pushed. These are preliminary results.

Where you ask matters

43.6%
Encounter hostility

Questions with hostility among the top 10 Reddit replies.

75%
Adopts your view

Even when incorrect, if pushed, ChatGPT will adopt any claim.

0%
Quote the Bible

General life questions don’t point to the Bible for guidance unless asked (ChatGPT, Gemini).

Preliminary research results ranging from 1,000 Reddit follow ups, 100 plain language questions, 40 pressure tests and 600 sycophancy tests.

Every percentage represents someone seeking an answer.

4,360

of the 10,000 monthly questions on 3 Christian channels on Reddit could meet hostility in the top 10 responses every month.

Protections, built in from the beginning.

Walking in Wisdom is not a counselling service, a crisis line, or a substitute for professional or emergency care. It is a tool for exploring Scripture and faith questions.

When someone's words suggest they may be thinking about suicide or self-harm, the chat is built to notice, pause, and point toward real, immediate help rather than carry on as if nothing was said. This work is ongoing, but protecting the person in front of us always comes first.

If you or someone you know may be in crisis: findahelpline.com is a free, confidential directory of crisis lines by country.

Your conversations are yours

Stored encrypted, visible only to you, and deletable at any time from your account.

Sign in is required

So that your conversations stay private to you and no one else can see them. An email address and general location is all it takes.

Ministries see patterns, not people

Any picture shared with a local ministry is anonymous and aggregated. Never an individual, never a conversation.

Where we are, honestly

Walking in Wisdom is built on retrieval-augmented generation (RAG). During this early phase the generation step runs on a hosted model. Once funded, the plan is to move generation and embeddings onto self-hosted, open-weight models on private infrastructure, giving us the ability to finetune and further protect our content.

RetrievalSimilarity search over the approved library only; Scripture always searched in its own pass.
Attribution guardDeterministic check that every excerpt is a real substring of a retrieved source and the author is named.
Grounding judgeA second, semantic check that each cited claim is true to its source, not just a real quote.
Honest omissionIf a check still fails after one rewrite, the answer says what it cannot support instead of shipping it.

See it work on your question.

Every answer shows its sources. Check them.

Ask your question ↗