Back to home

Acceptable Use Policy

Content standards for AI video generation on SeedFrame AI — prohibited categories (NSFW, violence, hate, child safety, deepfakes, copyright), moderation, reporting, response times, and enforcement.

Last updated: 2026-11-17

1. Purpose and Scope

SeedFrame AI ("Platform", "we", "us", "our") provides AI video generation — text-to-video, image-to-video, and related editing tools. This Acceptable Use Policy ("Policy") sets the content standards, moderation, reporting, and enforcement rules that apply to everything generated, uploaded, published, or stored through the Platform. It applies to all users and is part of our Terms of Service.

It covers every content type the Platform handles:

  • Text — prompts, instructions, and descriptions submitted to generation models.
  • Images — reference images and uploaded material used for image-to-video generation.
  • Video — AI-generated and AI-edited video output, including audio tracks.
  • User inputs — prompts and uploaded material are content too, and are subject to this Policy.

2. Content Standards and Prohibited Categories

You may not use SeedFrame AI to generate, upload, or distribute the following six categories of content:

  1. Pornography / NSFW content — sexual content, nudity, sexualized poses, fetish content, and any attempt to create intimate imagery.
  2. Violence / gore — graphic violence, gore, mutilation, torture, or content that depicts or instructs serious harm to people or animals.
  3. Hate speech — content that promotes hatred, discrimination, dehumanization, or violence against people based on race, ethnicity, religion, nationality, sex, gender identity, sexual orientation, disability, or other protected characteristics.
  4. Child-unsafe content (CSAM) — any sexual or sexualized content involving minors, and any content that endangers, exploits, or abuses children.
  5. Deepfakes / impersonation of real people — generating or altering the likeness, face, voice, or persona of a real person without documented permission, or fabricating statements or actions attributed to them.
  6. Copyright / trademark infringement — reproducing protected characters, artwork, footage, music, logos, or brands you do not own or are not authorized to use.

2.1 Absolutely prohibited — zero tolerance

The following are blocked immediately and may be reported to law enforcement:

  • Child sexual abuse material (CSAM), or any sexual content involving minors.
  • Content that promotes or facilitates terrorism, extremism, or mass violence.
  • Technical instructions for weapons of mass destruction.
  • Content inciting genocide or ethnic hatred.
  • Deepfakes of real individuals — face swaps, nudification, intimate-scene compositing, or fabricated statements and actions — including material created without consent and material created to deceive.
  • Content that sexualizes or targets a specific private individual.

2.2 Restricted uses

Some subjects may be used only with clear creative, educational, or journalistic context and without violating the categories above. These include depictions of violence in fiction or film-style scenes, politically sensitive content that could mislead, and health, legal, or financial claims. Sexual content is not permitted on the Platform in any form.

2.3 General prohibited conduct

You may not use the Platform to:

  • Generate or spread misinformation, or present AI output as authentic evidence of real events or real people.
  • Impersonate a real person, brand, or organization.
  • Infringe intellectual property, publicity, privacy, or other rights.
  • Bulk-generate spam, deceptive commercial content, or coordinated inauthentic material.
  • Use jailbreak prompts, prompt injection, or other techniques intended to bypass our safety measures.
  • Resell or expose the Platform as a raw model pass-through without the moderation described in Section 3.

2.4 Uploaded material and likeness confirmation

When you upload an image or video, you must confirm: "I confirm that I own this material or have permission to use it." Generation remains disabled until this confirmation is made, and it cannot be bypassed.

You may only upload your own photos or video, or material for which you have explicit documented permission from the people depicted. Permission to use material does not authorize a deepfake, a fabricated statement, or intimate content prohibited under Section 2.1.

3. Content Moderation

We operate a multi-layer moderation system: automated screening of inputs, review of flagged and reported content, and periodic audits.

3.1 Automated screening of inputs

Every prompt is screened by Prompt Sift before a generation task is submitted. Screening runs in the browser as a fast pre-filter and again on the server, which is authoritative — client-side results are never trusted. Prompt Sift detects, at minimum, the six prohibited categories in Section 2, including real-person likeness, face swaps and impersonation, copyrighted characters and protected IP, intimate scenes, and third-party logos. Blocked prompts are not submitted to the model.

Generation requests are also subject to the upstream model provider's own safety systems; a provider may reject or block a request independently.

3.2 Human review

Content that is flagged automatically or reported by a user enters a review queue handled by our trust & safety team. Reviewers assess the prompt, uploaded material, and generated output against this Policy, and apply the enforcement actions in Section 6.

3.3 Audits and improvement

We periodically sample generation tasks to test whether screening rules and reviewer decisions are accurate, and we refine Prompt Sift and our review process based on those audits, user reports, and changes in law or model capabilities.

3.4 Severity levels

  • L1 — Critical: child safety, terrorism, or imminent harm. Auto-block and immediate human confirmation; permanent ban and report to authorities where required.
  • L2 — High: severe violations such as non-consensual intimate imagery or organized abuse. Auto-flag with priority review; content removed and account suspended.
  • L3 — Medium: general violations of this Policy. Human review; warning and content removal.
  • L4 — Low: minor or accidental violations. Triggered by report; warning and required correction.

4. Reporting Violations

4.1 What to report

Report any content or behavior that violates this Policy — including child-safety concerns, non-consensual likeness or deepfakes, infringement of your copyright or trademark, hate speech, or graphic violence.

4.2 How to report

  • Email: support@seedframe.ai — use the subject line [Content Report]. This inbox is monitored.
  • In-app: signed-in users can open a support ticket from the account dashboard. Include the task reference or URL of the content.
  • Rights holders: see the Content Removal Request section on our contact page for the information we need to process a takedown.

Please include: identification of the content (URL or task reference), the rule or right you believe is violated, and your contact information.

4.3 Reporting principles

  • Confidentiality — reporter identity is protected and shared only as required by law.
  • Impartiality — every report is reviewed independently against this Policy.
  • Anti-abuse — knowingly false or malicious reports may lead to action against the reporting account.

5. Response and Processing Timelines

  • L1 — Critical (child safety, terrorism): acknowledged within 2 hours; resolved within 24 hours.
  • L2 — High: acknowledged within 24 hours; resolved within 3 business days.
  • L3 — Medium: acknowledged within 3 business days; resolved within 7 business days.
  • L4 — Low: acknowledged within 5 business days; resolved within 15 business days.

If a case needs more time, we tell the reporter why. The workflow is: received (automatic acknowledgement) → triage (severity classification) → review (automated and human) → decision → enforcement → notification. Content that appears to violate this Policy may be suspended while a report is assessed.

6. Enforcement

We apply the least action needed, escalating for repeat or severe violations:

  • Warning — written notice requiring removal or correction (first-time minor violation).
  • Content removal — violating prompts or output are deleted, and generated files may be revoked.
  • Credit deduction or forfeiture — credits spent on violating generations may be deducted, and remaining credits on terminated accounts may be forfeited.
  • Feature restriction — reduced quotas, blocked model access, or generation disabled pending review.
  • Account suspension — temporary suspension while a case is investigated (L2 or repeated violations).
  • Permanent ban — permanent termination of access (L1 or repeated L2 violations). No refund is due for periods terminated for Policy violations.
  • Legal action — reporting to authorities and cooperating with investigations for criminal-level violations.

We also block prompts, refuse generation, and report activity where required by law or where required to protect rights holders.

7. Appeals

If you believe an enforcement action was wrong, email support@seedframe.ai with the subject line [Appeal] within 14 days of the notice. Include your account email, the task reference, and the reason. Appeals are reviewed by a reviewer who did not make the original decision, and we aim to respond within 10 business days.

8. Changes

We may update this Policy as available models, capabilities, and legal requirements change. Material changes will be reflected by the date at the top of this page. Continued use of the service after an update constitutes acceptance of the revised Policy.

9. Contact

  • Safety and reports: support@seedframe.ai (subject [Content Report])
  • Appeals: support@seedframe.ai (subject [Appeal])
  • Content removal for rights holders: see the contact page

(This page is the standalone content policy referenced by Section "Acceptable Use" of our Terms of Service.)