Content Moderation Policy

How we moderate AI-generated content on our platform

2026/01/07

Introduction

This Content Moderation Policy explains how Wan 2.6 moderates AI-generated content to ensure a safe, legal, and respectful environment for all users.

Our Commitment

We are committed to:

  • Preventing generation of harmful, illegal, or inappropriate content
  • Protecting intellectual property rights
  • Maintaining a safe and inclusive platform
  • Complying with applicable laws and regulations

Automated Content Filtering

Input Filtering

Before content generation begins, we analyze:

  • Text prompts for prohibited keywords and concepts
  • Uploaded images for inappropriate content
  • Reference videos for policy violations

Prohibited inputs are automatically rejected with an explanation.

Output Filtering

After content generation, we scan for:

  • Violence, gore, or disturbing imagery
  • Sexually explicit content
  • Hate symbols or extremist imagery
  • Recognizable copyrighted characters or logos
  • Misleading deepfake characteristics

Flagged content is blocked from delivery to users.

Content Categories and Actions

Category 1: Automatically Blocked

Content depicting:

  • NSFW or sexually explicit material
  • Graphic violence or gore
  • Child safety violations
  • Hate speech or discriminatory content
  • Terrorist or extremist material

Action: Immediate blocking, no generation, account review

Category 2: Manual Review Required

Content that may involve:

  • Public figures or celebrities
  • Potential copyright concerns
  • Ambiguous policy violations
  • Complex creative projects

Action: Held for human review (typically within 4 hours)

Category 3: Allowed with Disclosure

Content that is:

  • AI-generated synthetic media
  • Creative transformations of user content
  • Educational or artistic projects
  • Commercial use with proper licensing

Action: Approved with appropriate disclosure labels

Model Source Transparency

Open Source Models

Wan 2.6 platform deploys and runs inference on open-source AI models including:

  • Models licensed under Apache 2.0
  • Models licensed under MIT License
  • Other permissively licensed open-source models

We do not own, develop, or train these models. We optimize and deploy publicly available models to provide our video generation service.

Model Attribution

For specific features:

  • Text-to-video: Based on open-source diffusion models
  • Image-to-video: Utilizing publicly available animation models
  • Audio synthesis: Deploying open-source audio generation models

User Data Protection

What We Do Not Do

  • We do NOT use your content to train AI models
  • We do NOT share your content with third parties without consent
  • We do NOT retain content beyond stated storage periods

What We Do

  • Process your inputs solely to provide requested services
  • Store generated content according to your subscription tier
  • Delete data permanently after retention period expires

Human Review Process

When Human Review Occurs

  • Flagged content from automated systems
  • User-reported violations
  • Random quality sampling (anonymized)
  • Account suspension appeals

Review Team Training

Our moderation team receives training on:

  • Platform policies and guidelines
  • Cultural sensitivity and context
  • Legal compliance requirements
  • Privacy and data protection

Review Timeline

  • Standard reviews: Within 4 hours
  • Complex cases: Within 24 hours
  • Appeals: Within 48 hours

Reporting and Appeals

Reporting Content

If you encounter policy violations:

  1. Note the content ID or URL
  2. Email support@hushgen.com with details
  3. Provide context for the violation
  4. We will investigate and respond within 48 hours

Appealing Actions

If your content was incorrectly flagged:

  1. Email support@hushgen.com with your case
  2. Include content details and reasoning
  3. Provide additional context if helpful
  4. Receive response within 48 hours

Transparency Reports

We publish quarterly transparency reports including:

  • Number of content reviews conducted
  • Categories of moderation actions taken
  • Appeals received and outcomes
  • Policy updates and improvements

Third-Party Integrations

When using third-party model providers:

  • We select providers with strong moderation policies
  • We apply additional filtering on top of provider safeguards
  • We maintain the same standards across all models

Children's Safety

Special protections for minors:

  • Users must be 13+ to use the platform
  • Enhanced filtering for content involving children
  • Strict prohibitions on inappropriate content involving minors
  • Immediate reporting of CSAM to authorities

Continuous Improvement

We regularly:

  • Update filtering algorithms based on emerging threats
  • Review and refine policy guidelines
  • Train moderation teams on new patterns
  • Incorporate user feedback into improvements

We comply with:

  • Digital Services Act (EU)
  • Online Safety regulations
  • DMCA and copyright laws
  • Data protection regulations (GDPR, CCPA)

Contact Information

For moderation questions or reports:

  • Email: support@hushgen.com
  • Response time: 24-48 hours
  • Emergency issues: Clearly mark subject line as "URGENT"

Last Updated: January 7, 2026