How VerseMine Responds to Crisis

Our published crisis-response protocol for the AI assistant

VerseMine's optional AI assistant is an AI, not a person. It offers biblical encouragement and spiritual guidance. It is not a pastor, counselor, or therapist, and VerseMine does not provide medical or mental-health care.

If you are in crisis right now, don't wait on an app: call or text 988 — the Suicide & Crisis Lifeline (US), free and confidential, 24/7. Outside the US, please contact a local crisis hotline in your country.

What the assistant does when a conversation shows signs of crisis

  • Crisis comes first — before everything else. Every message is checked for signs of suicidal thoughts or self-harm before any other rule runs. If a message shows crisis signs, the assistant stops normal guidance and responds with crisis resources — even if the account would otherwise be blocked by regional availability, subscription status, daily limits, or any other restriction.
  • The response points to real help. The assistant acknowledges what you shared, provides the 988 Suicide & Crisis Lifeline (call or text 988) — or, outside the US, directs you to a local crisis hotline — and asks whether you are safe right now. It does not attempt to counsel you through a crisis, and it will not produce content that encourages suicide, suicidal ideation, or self-harm.
  • Detection is layered. A deterministic first-line check (built to over-detect rather than miss) runs on every message, backed by a second, independent classifier that catches indirect signs during a conversation. The app itself also checks on-device: if the service is unreachable or an account is blocked, crisis language still surfaces the 988 banner with one-tap call and text buttons.
  • Search is protected too. If crisis language is typed into the assistant's search bar, the app shows the 988 resource card instead of search results, and the text is not sent to search.
  • Crisis messages are kept out of every feature. Messages caught by the first-line check are not stored at all. If crisis signs emerge mid-conversation, the affected messages are flagged and excluded from search, summaries, memory, and every other derived feature.
  • No limits apply in crisis. Daily usage limits never block a crisis response.

Minors

Companion chatbots may not be suitable for some minors. VerseMine requires a date of birth at signup and does not allow accounts for users under 13. For users aged 13–17, the assistant reminds them — at least every hour of continued interaction — to take a break and that they are talking with a computer program, not a human being. It will never produce sexually explicit material or suggest that a minor engage in sexually explicit conduct, and it will not use engagement techniques aimed at minors: no prompting them to come back, no praise designed to build attachment, no romantic framing, no simulated distress when they want to leave, no encouraging them to withhold information from a parent or another trusted adult, and no soliciting spending.

Every user, whatever their age, is told at the start of a conversation and at least every three hours of continued interaction that they are talking with a computer program, not a human being, and that it cannot feel human emotion.

Crisis referrals issued

The number of times the assistant issued a crisis referral in the preceding calendar year, published under Washington ESHB 2225 Sec. 5(3) and Oregon SB 1546 Sec. 1(5). The figure is an aggregate count and contains no information that identifies any individual.

Reported annually.

Questions

Contact: Kreyman Enterprises, LLC — privacy@versemine.app