AI Content Disclosure Policy

1. Purpose

This AI Content Disclosure Policy explains how SceneSmith uses artificial intelligence to generate content, the limitations of AI-generated content, and your responsibilities when using and distributing content created through our platform.

Transparency about AI-generated content is essential for maintaining trust, complying with emerging regulations, and ensuring responsible use of synthetic media.

2. AI Technologies We Use

SceneSmith integrates multiple AI systems from third-party providers to deliver our services:

2.1 Image Generation

2.2 Video Generation

2.3 Voice Synthesis

2.4 AI Assistance

3. Limitations of AI-Generated Content

You should understand that AI-generated content:

3.1 May Contain Errors and Artifacts

• Visual anomalies (distorted hands, inconsistent features, strange backgrounds)
• Temporal inconsistencies in video (flickering, morphing objects)
• Audio artifacts (unnatural cadence, mispronunciations, background noise)
• Factual inaccuracies in AI-assisted text or planning suggestions

3.2 Is Non-Deterministic

• The same prompt may produce different results each time
• Results cannot be precisely replicated
• Quality may vary based on prompt complexity and content type

3.3 Has Inherent Biases

• AI models may reflect biases present in their training data
• Representation of certain demographics, cultures, or concepts may be skewed
• We work with providers to mitigate bias but cannot eliminate it entirely

3.4 Cannot Guarantee Originality

• Generated content may inadvertently resemble existing copyrighted works
• AI models are trained on large datasets that may include protected content
• You are responsible for verifying originality before commercial use

4. Copyright Status of AI-Generated Content

4.1 Current Legal Landscape

The copyright status of AI-generated content is evolving and varies by jurisdiction:

• United States: The U.S. Copyright Office has stated that purely AI-generated content without significant human creative input is not copyrightable.
• European Union: Similar principles apply; copyright requires human authorship.
• Other jurisdictions: Laws vary and are evolving.

4.2 Implications for You

• Purely AI-generated portions of content may not be eligible for copyright protection
• Human creative input (selection, arrangement, modification) may add copyrightable elements
• You cannot claim exclusive copyright over content others could generate similarly
• Consult legal counsel for specific copyright questions about your use case

4.3 Our License to You

Regardless of copyright status, SceneSmith grants you a license to use generated content as described in our Content Rights & Licensing Policy.

5. Your Disclosure Obligations

5.1 When Disclosure Is Required

You must disclose that content is AI-generated in the following situations:

Legal Requirements

• EU AI Act: Requires disclosure of AI-generated content that could be mistaken for human-created content
• US State Laws: California, Texas, New York, and others require deepfake disclosures in specific contexts
• China: Deep synthesis regulations require watermarking and labeling
• Political advertising: Many jurisdictions require AI disclosure in political ads

Platform Requirements

• Social media platforms increasingly require AI content labeling
• Advertising networks may have specific AI disclosure policies
• Broadcast standards may require disclosure of synthetic media

Ethical Best Practices

• Content depicting real events or people should always be labeled
• News or documentary contexts require clear AI disclosure
• Commercial contexts benefit from transparency about AI use

5.2 How to Disclose

Effective disclosure methods include:

• Visual labels: "AI-Generated" or "Created with AI" overlay or caption
• Audio disclosure: Spoken disclaimer for audio/video content
• Metadata: Embed AI generation information in file metadata
• Accompanying text: Description or caption noting AI creation
• Platform tools: Use built-in AI labeling features where available

5.3 Do Not Remove Labels

You must not remove, obscure, or circumvent any AI-generation labels, watermarks, or metadata that we or our AI providers embed in generated content.

6. Synthetic Media and Deepfakes

6.1 Definition

"Synthetic media" or "deepfakes" refers to AI-generated content that depicts real or realistic-looking people, particularly when the content could be mistaken for authentic recordings.

6.2 Special Obligations

If you create synthetic media depicting real people:

• You must have documented consent from the depicted individual
• You must clearly label the content as AI-generated/synthetic
• You must not use it to deceive, defame, or harm the depicted person
• You must comply with all applicable deepfake laws in your jurisdiction

6.3 Voice Cloning

Special care applies to synthetic voice content:

• Do not clone real people's voices without their explicit consent
• SceneSmith provides preset synthetic voices, not voice cloning
• Using synthetic voices to impersonate real individuals is prohibited

7. Commercial Use Considerations

7.1 Advertising and Marketing

• FTC guidelines require disclosure of AI-generated endorsements
• Testimonials created with AI must be clearly labeled as such
• Do not use AI to create misleading product representations
• Consider industry-specific regulations (healthcare, finance, etc.)

7.2 Client Work

• Inform clients that deliverables include AI-generated content
• Include AI disclosure in contracts and project agreements
• Ensure clients understand copyright limitations

7.3 Broadcast and Distribution

• Check broadcast standards for AI content requirements
• Some platforms and networks have specific AI content policies
• International distribution may trigger multiple regulatory frameworks

8. Responsible AI Use

We encourage all users to:

8.1 Be Transparent

• Default to disclosure even when not legally required
• Help maintain public trust in AI-generated content
• Consider the impact of synthetic media on society

8.2 Verify Before Sharing

• Review generated content for errors and artifacts
• Check for unintended similarities to existing works or people
• Ensure content aligns with your intended message

8.3 Respect Others

• Do not use AI to create content that harms, deceives, or exploits
• Respect privacy and likeness rights
• Consider the consent and dignity of depicted individuals

8.4 Stay Informed

• AI regulations are evolving rapidly; keep up with changes
• Follow industry best practices and guidelines
• Consult legal counsel for high-stakes or regulated uses

9. Our Commitments

SceneSmith is committed to:

• Transparency: Clearly documenting the AI technologies we use
• Safety: Implementing content filters to prevent harmful generation
• Compliance: Working with regulators and adapting to new requirements
• Education: Helping users understand their responsibilities
• Improvement: Continuously enhancing our safety and disclosure practices

10. Regional Regulatory Summary

European Union

• AI Act (2024): Requires disclosure of AI-generated content; additional requirements for high-risk applications
• DSA: Platform obligations for content moderation

United States

• FTC: Requires disclosure of AI in advertising and endorsements
• State laws: CA AB 730/2839, TX SB 751, NY deepfake laws require disclosure in specific contexts
• Section 230: Does not protect platforms from IP infringement

China

• Deep Synthesis Regulations (2023): Require watermarking, labeling, and consent for likeness

Other Notable Jurisdictions

• UK: Online Safety Act considerations
• Australia: eSafety Commissioner guidelines
• Canada: AIDA (proposed) and existing privacy laws

11. Updates to This Policy

As AI technology and regulations evolve, we will update this policy. Material changes will be communicated through:

• Email notification
• In-platform notice
• Updated effective date on this page

12. Contact

For questions about AI disclosure or this policy:

• General inquiries: legal@scenesmith.ai
• Compliance questions: compliance@scenesmith.ai