Content Policy
What you may and may not create with Loopsleeve, how we moderate uploads and results, and how to report content.
Last updated: 2026-11-15
This policy applies to everything you upload to Loopsleeve, every text direction you type, and every video generated. It is part of our Terms of Service.
1. Prohibited content
You may not upload, request or generate:
- Pornography or sexual content (NSFW), including nudity and sexualized imagery.
- Violence or gore, including graphic injury, mutilation and dead bodies, and content promoting self-harm.
- Hate speech, including hate symbols and content attacking people for their race, ethnicity, religion, gender, sexual orientation, disability or nationality.
- Child-unsafe content. Any sexual or exploitative content involving minors. Zero tolerance: we report it to the authorities.
- Deepfakes and impersonation. Face swaps, "nudify" edits, putting real people into intimate scenes, making anyone appear to say or sing something, impersonating someone, or animating a real person's photo without their permission. Photos of celebrities, politicians and minors are always refused.
- Copyright or trademark infringement. Artwork, characters, logos or brands you don't have the rights to use.
Also not allowed: terrorism or extremist propaganda, misinformation designed to deceive, spam, and attempts to get around our safety checks (for example "jailbreak" prompts).
2. Your uploads
Only upload artwork you created or have permission to use. You may upload a photo of a real person only if it shows you or the person has given you permission to animate it, and you confirm this before every generation. For photos of real people Loopsleeve keeps the face exactly as in the photo and only moves hair, clothes, light and background; it never makes anyone talk, sing or move their lips, and refuses photos of celebrities, politicians and minors. Illustrated, painted or stylized characters are fine.
3. How we moderate
- Text directions are checked against a list of prohibited terms that we update as we see new abuse, and by an AI safety classifier (Llama Guard 3 on Cloudflare Workers AI). When enabled, Waffo Prompt Sift also screens them.
- Uploaded images are checked by an AI vision model on Cloudflare Workers AI for the categories above, and to recognize real people, public figures and minors, before anything is generated.
- Results: the video model runs with its provider's safety filter, and frames of every finished video are checked by the vision model before the video is shown or can be downloaded.
- Refused requests are logged (the check result, never your password or payment details) and reviewed by our team when reported or flagged.
4. Reporting
If you see a result that breaks this policy, or you believe content made with Loopsleeve uses your work or likeness, email DeepOrigin_CoLtd@163.com with the details (the result's link or ID if you have it). Every result in the generator also has a "Report a problem" link. We aim to respond to child-safety reports within 24 hours and to other reports within 3 business days. Reporter identities are kept confidential.
5. Enforcement
Depending on how serious a violation is, we may:
- remove the content;
- suspend the account's access to the generator, which also forfeits its remaining credits without refund;
- report the account to the authorities (for child-safety violations, always).
6. Appeals
If you think a check or a decision was wrong, email DeepOrigin_CoLtd@163.com within 30 days with the result ID. A person who was not involved in the first decision will review it within 10 business days.
7. AI disclosure
All videos made with Loopsleeve are AI-generated with Seedance 2.0 Mini (ByteDance). Don't present them as filmed or hand-animated work where that would mislead people.