Content Moderation Policy

Girlfriendly AI operates a layered moderation program covering everything created on the platform: prompts, chats, generated images and videos, character definitions, and profile content. All content on the service is AI-generated, and moderation applies before, during, and after generation.

Automated moderation

Automated classifiers screen user prompts before generation and screen generated outputs before delivery. Requests and outputs that indicate prohibited content — including any depiction of minors, non-consensual scenarios involving real people, or other illegal material — are blocked automatically. Automated systems also flag borderline content for human review rather than allowing it by default.

Human review

Automated systems are backed by trained human moderators. In particular:

• every character submitted for public visibility is reviewed by a person before it is published
• content flagged by classifiers, user reports, or monitoring sweeps is escalated to human review
• moderator decisions on serious categories are made by people, not solely by automation

Ongoing monitoring

Moderation does not end at publication. We continuously monitor public areas of the platform, re-scan published content as our classifiers improve, and run periodic sweeps of existing characters and media. Content that was previously acceptable may be removed if standards, the law, or our understanding of it changes.

Enforcement ladder

Violations are handled proportionately to their severity and the account's history. Typical escalation is:

1. removal or restriction of the violating content
2. a warning recorded against the account
3. restriction of specific features, such as public publishing or generation
4. temporary account suspension
5. permanent account termination

Severe violations — including any child sexual abuse material, credible threats, or content facilitating trafficking or exploitation — skip the ladder and result in immediate termination.

Zero tolerance: child sexual abuse material

Girlfriendly AI maintains absolute zero tolerance for child sexual abuse material and any sexualized depiction of minors, whether real, AI-generated, or stylized. Confirmed material is removed immediately, the responsible account is permanently terminated, relevant records are preserved, and the matter is reported to the National Center for Missing & Exploited Children (NCMEC) and to law enforcement as required by law. See our Underage Policy for full details.

Reports and appeals

Anyone can report content through our Report Abuse / Takedown Policy page or by emailing [email protected]. Users subject to an enforcement action may appeal as described in our Content Removal Policy; appeals are reviewed by a person not involved in the original decision.

This policy works together with our Pre-Screening and Post-Screening Policy, Complaints Policy, and Terms of Service.