Last updated: 27 July 2026 (version 2026-07-27)
Mentor is an AI reflective-coaching application. This page publishes our safety protocol as required by California Senate Bill 243 (§22602(b)(2)) on companion chatbots. It describes how we detect signals of suicidal ideation or self-harm during a conversation, how we respond, and where to find real help.
A dedicated safety layer runs on every message before the coach replies. It looks for language indicating suicidal ideation, self-harm intent, or acute crisis. When such signals are detected:
This behavior takes priority over every other coaching rule and cannot be disabled by the user or overridden by conversation context.
The resources below are surfaced by the app when relevant, and are always available directly:
| Country | Service | Contact |
|---|---|---|
| United States | 988 Suicide & Crisis Lifeline | Call or text 988 · 988lifeline.org |
| France | Numéro national de prévention du suicide | Call 3114 (free, 24/7) · 3114.fr |
| United Kingdom | Samaritans | Call 116 123 · samaritans.org |
| European Union | Common emergency number | Call 112 · European suicide helplines: findahelpline.com |
| Canada | Talk Suicide Canada | Call 1-833-456-4566 · text 45645 |
| Australia | Lifeline Australia | Call 13 11 14 · lifeline.org.au |
| Other | International directory | findahelpline.com |
Mentor does not diagnose, treat, cure, or prevent any medical or mental-health condition. Conversations with Mentor are not a substitute for care from a licensed professional (physician, psychiatrist, psychologist, therapist, counselor).
If you are receiving mental-health treatment, please continue that treatment. Do not stop or adjust prescribed medication or a therapeutic plan based on a conversation with Mentor.
Mentor is intended for adults (18+ only). Even among adults, Mentor may not be suitable for everyone — for example, individuals in active psychiatric crisis, people whose clinician has advised against using self-directed reflection tools, or people who find conversational AI distressing. If Mentor makes you feel worse, stop using it and speak with a professional.
Mentor may not be suitable for some minors. Access is restricted to adults 18 and older; see our Terms of Service.
Messages that trigger the safety layer are stored like any other message under our normal retention policy (see Privacy Policy). They are used only to operate the safety response itself and to improve detection quality. They are never sold, shared with advertisers, or used for marketing.
Any team review of conversation content passes through our anonymization pipeline (regex + LLM classifier stripping personal identifiers). This is a discipline for human review, not an absolute anonymization guarantee — contextual details may still allow re-identification by someone who already knows the user. We disclose this limitation openly.
If you believe our safety protocol failed in a specific conversation, or if you have a concern about a user's safety, email contact@get-mentor.com. We read this address and respond within 3 business days.
This protocol is reviewed at least once per year and every time a material change is made to the safety layer. Version history is tracked in the "Last updated" field at the top of this page.