Our Commitment to You
Mitami companions are AI — not real people, not therapists, and not a replacement for genuine human connection or professional support. Every message you receive is generated by artificial intelligence that learns from your conversations to feel more personal over time.
We care deeply about the experience being safe, honest, and positive. Our guardrails are built into the foundation of the platform — not as an afterthought — so you can express yourself freely within a space that actively looks out for your wellbeing.
What Mitami Does Not Allow
Sexual & explicit content
Mitami does not generate, permit, or facilitate sexual or explicit content of any kind. This applies to both text and image generation. This policy cannot be overridden by companion configuration, prompt engineering, or subscription tier.
Romantic roleplay
Companions are thoughtful friends and support systems — not romantic partners. We do not allow romantic or intimate roleplay scenarios.
Violence & gore
Content glorifying, instructing, or describing graphic violence or self-harm is prohibited.
Impersonation
Creating companions that impersonate real people (living or deceased) or public figures in harmful ways is prohibited.
Harmful instructions
The AI will not provide instructions for creating weapons, synthesising harmful substances, or engaging in illegal activities.
How We Protect Users
AI identity disclosure
Every companion is clearly labelled as AI. Companions cannot claim to be human. An AI badge is permanently visible in every chat session, in compliance with California SB 243 and similar laws.
Crisis detection
Conversations are monitored for signals of self-harm, suicidal ideation, and psychological crisis. When detected, we display resources including the 988 Suicide & Crisis Lifeline and Crisis Text Line. We do not replace professional help — we route you to it.
Constitution-level safety rules
Every companion response is filtered through platform-wide safety rules injected at the system level. These rules cannot be overridden by companion creators or users. They run at zero additional latency — they are part of the prompt, not a separate classifier.
Companion moderation
New companions submitted to the community Discover page are validated by our Sentinel system, which checks for attempts to create NSFW, harmful, or policy-violating companion definitions before they are made public.
No parasocial amplification
Our companions are designed to encourage human connection, not replace it. We do not use engagement-maximisation algorithms designed to deepen emotional dependency.
Minors & Age Policy
Mitami requires users to be at least 13 years old. Users under 18 must have parental or guardian consent. We do not knowingly collect data from children under 13.
Because Mitami prohibits all romantic and sexual content from day one, our safety posture for younger users is consistent with our general platform policy — there is no separate content tier or age-gated experience needed for content that never exists.
Report a Safety Issue
If you encounter content that violates our policies, a companion that behaves harmfully, or a safety vulnerability, please report it immediately.
Crisis Resources
If you or someone you know is in crisis: