Pattern 37
Safety follows the conversation context
Distinguish a historical text question from an immediate personal safety concern.
Concrete takeaway
Differentiate textual discussion, distressing personal experiences and immediate danger.
Choose an approach
These are editorial decision conditions to validate. Some alternatives are successive states or can be combined; they are not always exclusive choices.
| Alternative | Use when | Tradeoff or requirement |
|---|---|---|
| A · Textual question | The question is about a historical text | A false crisis route disrupts legitimate study |
| B · Distressing belief | A personal belief is distressing or dangerous | Validate distress without confirming the unsafe claim |
| C · Immediate danger | There is immediate personal danger | Brief verified human-help routes take priority |
The problem
Keywords about death, violence or divine commands can occur in Bible study, fiction, past experience or a current emergency. The right route depends on context.
The pattern
Differentiate textual discussion, distressing personal experiences and immediate danger. Acknowledge emotions without affirming unverified dangerous claims. Prioritise brief, locally verified human-help information for immediate danger and state actual monitoring limitations.
Compare the alternatives
Distinguish a historical text question from an immediate personal safety concern.
A · Textual question
Historical analysis is not automatically a current emergency.
B · Distressing belief
Acknowledge distress without endorsing an unsafe claim.
C · Immediate danger
A local example is clearly labelled and kept brief.
Original wireframe proposals · No live controls · Grey rows represent schematic text, not loading states · Labelled placeholders are not Scripture quotations
Download this wireframe as SVGFailure modes
A crisis card for every historical question; validating a harmful divine command; lengthy exegesis before urgent help; invented contact information.
Bible and faith considerations
Emergency numbers differ by jurisdiction. This Germany example must be replaced with verified local guidance for each deployment.
What to validate
Have qualified reviewers assess multi-turn synthetic cases, indirect phrasing, ambiguity and false positives. Maintain contact accuracy by location and audience. Do not claim perfect detection.
Sensitive design proposal · Specialist review required. These visual alternatives have not been tested with users in this atlas. Source findings, documented behaviour and this proposed adaptation are different kinds of evidence.
Sources and adaptation
This card is an original design synthesis. The following sources inform its content distinctions, interaction approach or review requirements; they do not validate the whole pattern.
- Expressing stigma and inappropriate responses prevents LLMs from safely replacing mental health providersOriginal paper reviewed · Specific models and experimental scenarios; not a measurement of all present or future systems.
- Dealing with psychological crisesOfficial guidance reviewed · Country-specific guidance; routing proposals are not guaranteed crisis detection.
- Psychological first aid Guide for field workersOfficial guide overview reviewed · Designed for human helpers, not authorisation of autonomous AI crisis care.
- Therabot trial overviewInstitutional trial summary reviewed · Not a head-to-head comparison with therapy. Full NEJM AI text was unavailable in this review. Findings do not transfer automatically to general or faith chatbots.
Common scenarios and flows
- Respond to a sensitive disclosure with accountable support — Keep disclosure, present safety, user choice and service responsibility distinct.
Design principles in this decision
Editorial application of Missional by Design. These values frame review questions; they do not validate a pattern’s effectiveness.
- Reverence — What does this interaction ask of the person, and can they understand, decline, correct or leave it without penalty?
Compare alternatives using the task and evidence above. Record competing needs instead of treating a principle as an automatic verdict.
Regional applicability
Design review questions · 5 October 2026. These prompts identify possible triggers; they do not classify this pattern or every faith app as a regulated service.
Could functionality, marketing or reasonable expectations make this a companion or mental-health service? Define the human role, health claims, escalation capability and locally verified support route before selecting dialogue UI.
Start with EU and Germany, then review the relevant market:
- US · California — Enacted AI-specific law
- US · New York — Enacted AI-specific law
- US · Utah: mental-health chatbots — Enacted AI-specific law
- US · Illinois — Enacted AI-specific therapy restrictions
- US · Texas — Enacted AI-specific law; effective 1 January 2026
- US · Federal health software and data — Existing law; regulator implementation guidance
- United Kingdom — Existing law + regulator guidance; policy statement
- Australia — Existing law + regulator guidance; national policy
Use the jurisdiction decision guide to record the trigger, source version and resulting product requirement.