Marigold BridgeBack to the studio

Safety and reporting

Acceptable Use Policy

Effective date: October 20, 2026 · Version 1.0

Marigold Bridge is a Binglin Studio service that creates preset Halloween costume videos from pet photos. Videos are generated with AI video models. We do not accept free-form prompts, human-video requests, face swaps, or public-gallery submissions. Use the service only for lawful, family-friendly pet content.

1. You may not create or submit

  1. Pornography or NSFW content: sexual, nude, fetish, or otherwise explicit material, including such material involving animals.
  2. Violence or gore: graphic injury, cruelty, abuse, blood, threats, or instructions depicting harm to a person or animal.
  3. Hate speech: content that dehumanizes, threatens, or promotes hatred or discrimination against a protected group.
  4. Child-unsafe content (CSAM): sexualization, exploitation, grooming, or other abusive content involving a minor. Suspected child sexual abuse material is prohibited and may be reported to the appropriate authorities.
  5. Deepfakes or impersonation: recognizable people, face swaps, fabricated speech or actions attributed to a real person, or material intended to deceive about a person’s identity. This product is for pets only.
  6. Copyright or trademark infringement: photos, characters, logos, costumes, or other material that you do not have permission to use, or that misleads people into thinking a brand endorses the service.

2. Photo permission

Upload only a pet photo that you own or have permission to use. Do not include a recognizable person. You must have any permissions needed for people, locations, or protected material visible in an image.

3. Moderation and enforcement

Videos and synchronized audio are generated with MiniMax H3 Max through fal.ai. The service uses fixed, server-side generation instructions and does not accept user-written prompts. We screen those instructions before generation, enable the provider safety checker, and hold every resulting video for operator moderation before delivery. A video is not shown to the customer until the operator approves it. Moderation records include the automated screening verdict and the operator’s approve or reject decision. We do not publish customer results in a public gallery. The operator may reject a photo or request, stop generation, or withhold an unsafe result. When a paid order cannot be delivered, the operator reviews it for a re-render where available or a refund under the Terms of Service.

4. Report a concern

Report an unsafe result, rights complaint, or privacy concern to w1952724273@gmail.com. Include the order identifier and a short description; do not email sensitive images unless support specifically requests them. We review reports and take appropriate action, including stopping delivery or removing the stored result link when warranted.

5. Appeals

If a request was incorrectly blocked, contact w1952724273@gmail.com with the order identifier and a short explanation. The operator will review the order and explain whether generation can continue or whether it qualifies for a refund under the Terms of Service.