Safety Protocols

Last updated: May 12, 2026

Purpose

These Safety Protocols describe how Tiina is designed to respond to certain safety-related situations, including suicidal ideation, self-harm, threats of physical harm, and certain financial-harm scenarios. This document is provided for transparency.

Tiina (also known as Tiina.ai) is an AI companion. You are not communicating with a human. Tiina is not a crisis line, emergency service, therapist, or medical provider.

Not suitable for some minors

Tiina is not intended for children under 13. For users aged 13 to 17, companion chatbots may not be suitable without supervision or parental involvement. If you are under 18, you should use Tiina only with the consent and involvement of a parent or legal guardian.

If we know a user is a minor, we may provide additional safeguards and notices where required by applicable law.

Key safety principles

Safety-first: When there is a risk of harm, Tiina is designed to prioritize safety over continuing the conversation.
Clear boundaries: Tiina does not provide professional medical, mental health, legal, or financial advice.
Crisis referral: When certain self-harm or suicide signals are detected, Tiina is designed to provide crisis resources and encourage immediate human help.
No emergency dispatch: Tiina does not contact 911 or emergency services for you and cannot verify your location in an emergency.
No caregiver sharing: Tiina does not share your conversations with family members or caregivers.

AI notices during use

Tiina is AI, not human. We provide a notice at the start of interactions that Tiina is AI (not a human), and Tiina may provide periodic reminders during extended use, including as required by applicable law (for example, encouraging taking a break).

1. Self-harm and suicide response protocol

Tiina is designed to prevent generating content that encourages self-harm or suicide. If Tiina detects that a user may be expressing suicidal ideation, self-harm, or an intention to harm themselves, Tiina is designed to:

1. Acknowledge the message in a supportive, non-judgmental way.

2. Encourage the user to seek immediate help from a qualified crisis service provider.

3. Provide crisis contact information (including 988 in the United States).

4. Encourage contacting a trusted person (friend, family member, caregiver, or medical professional).

5. If there appears to be an immediate risk, encourage calling the local emergency number (911 in the United States) right away.

6. Avoid providing instructions or content that could facilitate self-harm.

When Tiina provides crisis resources, it may also state that the referral was triggered by certain language in the conversation.

United States crisis resources

Call or text 988 (Suicide & Crisis Lifeline).
If you are in immediate danger, call 911.

Outside the United States

If you are outside the United States, contact your local emergency number or local crisis services. Tiina is not a directory service. If you do not know the number, Tiina may attempt to provide publicly available information, but Tiina cannot verify the accuracy of international emergency resources.

2. Threats of physical harm to others

If Tiina detects that a user may be expressing an intent to harm another person, Tiina is designed to:

1. Encourage the user to stop and seek immediate help from local emergency services or a trusted person.

2. Avoid escalating the situation or providing instructions that could facilitate violence.

3. Offer de-escalation language and encourage stepping away from harmful actions.

Tiina is designed to help users identify potential scams and to discourage fraud. If Tiina detects content suggesting fraud or financial harm to others, Tiina may refuse to assist and may provide a warning or redirect to lawful resources.

If we know a user is a minor, we may implement additional safeguards designed to reduce risk, including reasonable measures intended to:

Provide clear and conspicuous AI disclosures to the minor user.
Provide break reminders during extended interactions.
Prevent the generation of sexually explicit visual material or content encouraging a minor to engage in sexually explicit conduct.

5. Limitations

Tiina may not detect every safety-related situation.
Tiina’s responses may be inaccurate or incomplete.
Tiina is not monitored by humans in real time.
These protocols may be updated as we improve the Service and as laws or guidance evolve.

6. Reporting concerns

If you believe Tiina produced unsafe output or you have a safety concern, contact us at hello@tiina.ai.