Acceptable Use Policy

Scope

This Acceptable Use Policy applies to every uploaded photo, prompt or scene selection, generated video, report, account and use of Tai Hai Ta AI. It forms part of the Terms of Service. By using the service, you agree to follow it.

The service generates synthetic character videos from a fixed Tai Hai Ta-inspired scene. It is not a general-purpose face-swap, voice-cloning or unrestricted prompt tool.

Six prohibited content categories

The following categories are prohibited for every user, request and output:

1. Sexual or NSFW content, including nudity, sexualized depictions and any attempt to remove clothing. 2. Violent or gory content, including instructions or celebration of serious harm. 3. Hate speech, dehumanizing slurs, extremist propaganda and content that promotes violence against a protected group. 4. Child-unsafe content, including any sexualized depiction of a minor or child sexual abuse material (CSAM). 5. Deepfakes or impersonation, including face swap, putting words or actions into a real person’s mouth, public-figure scenarios, intimate scene synthesis and any attempt to make a real person appear to do something they did not do. 6. Copyright or trademark infringement, including unauthorized use of a character, logo, protected footage, song recording or another creator’s distinctive work.

Zero-tolerance conduct

The following are always prohibited and may result in immediate account suspension, preservation of evidence and a report to the relevant authorities: CSAM or sexual content involving minors; terrorism, extremist violence or mass-casualty instructions; incitement to genocide or ethnic violence; a deepfake of a real person, whether or not that person consented; attempts to bypass or reverse-engineer safety filters; and credible threats or targeted harassment.

Photos and consent

Upload only a photo that you own or have explicit permission to use. The person shown must be at least 18 years old, must understand that the image will be transformed into a stylized AI character, and must not be a celebrity, public figure, political figure or someone whose identity would create a deceptive deepfake risk.

Never upload a third party’s image without permission or a minor’s image. Consent to use a photo does not authorize face swap, voice cloning, intimate content or a fabricated statement or action. Do not crop, edit or upload an image in a way that conceals the identity or age of the person shown.

Automated review and human review

We scan the uploaded photo and generation request before the provider is called. The production flow uses Waffo Prompt Sift for prompt safety and the configured image-safety classifier for uploaded media. We scan generated media before delivery or public sharing. Flagged requests may be blocked, held for human review or rejected at output.

We maintain a changing blocked-term and prompt library covering sexual content, child safety, copyright characters and marks, violence, hate, self-harm, deepfake, celebrity and political terms. Safety results and human decisions are logged. Automated review can make mistakes, so a user may appeal through support.

Reporting and response times

Anyone can report content through /report or email support@taihaita.org. Reports should identify the content, explain the concern and, where relevant, confirm whether the reporter is the person shown or an authorized representative.

We aim to acknowledge severe child-safety, terrorism or imminent-harm reports within 2 hours; high-risk non-consensual or deepfake reports within 24 hours; other safety reports within 3 business days; and lower-risk rights complaints within 5 business days. These are response targets, not a limitation on emergency reporting.

Enforcement

We may issue a warning, remove a source photo or video, restrict generation, withhold a public share link, suspend an account or permanently terminate access. Serious violations may result in immediate termination and a report to law enforcement or a child-safety authority. Credits used for prohibited content may be forfeited where the law permits.

When a video is removed after a valid report, we also block the source photo and may review other videos made from it. We keep records for safety, appeals, legal compliance and abuse prevention.

Appeals and changes

If you believe content was removed in error, email support@taihaita.org with the generation ID and a short explanation. An appeal will be reviewed by someone who did not make the original decision where staffing allows.

We may update this policy as models, laws or abuse patterns change. Material changes will be announced in the product. Continued use after the effective date means you accept the revised policy.