Sora disclosure guide
Does content created with Sora need AI disclosure?
Direct answer
Realistic Sora video commonly needs disclosure because it is generated video designed to look coherent and authentic. Use the platform's altered-content control and keep the embedded provenance intact. State which scenes or components were generated, verify the final export, and review product claims, depicted events and likeness permissions independently of the AI label.
Applies when
- Generated news-style footage of an event that never happened — Disclosure required.
- Real product image animated into synthetic action — Disclosure recommended.
Doesn't apply when
- Clearly fantastical animation — No disclosure identified.
Worked decision
Example
Generated news-style footage of an event that never happened
Decision: Disclosure required
Assess the final output rather than the tool name alone.
Video generated with Sora. The depicted event is fictional.
Next steps
Required action
- Record exactly how Sora was used and which component of the final asset it affected.
- Verify the final export instead of assuming its provenance status: C2PA documented for Sora videos.
- Apply the destination platform’s native label and precise audience-facing wording where the resulting content triggers them.
Publishing layer
Platform differences
Detection and native labels
YouTube's altered-content workflow is directly relevant to realistic Sora video.
Policy threshold
TikTok requires AIGC labelling for realistic generated video and audio.
Publishing workflow
Meta can apply AI info based on declaration or technical signals.
Evidence
Source
Freshness and accountability
Last verified
Last reviewed 2026-09-12 · Current
Run this exact scenario
Sora and a representative content scenario are prefilled. Confirm the actual AI operation, affected component and destination.
Tool → type of AI usage
- Text-to-video generation
- Image-to-video generation
- Video editing and extension
- Generated audio and consent-controlled likeness features
Does Sora generate C2PA or other provenance?
documented
C2PA documented for Sora videos
OpenAI states that Sora videos embed C2PA and include visible and invisible provenance signals; many outputs also carry a visible moving watermark. Preserve these signals, but do not treat them as a substitute for platform or Article 50 disclosure.
EU implications
- Realistic generated people, places, objects or events can meet the Article 50 deepfake definition.
- Uploaded likenesses add consent and rights requirements.
- C2PA helps prove origin but does not prove the scene is truthful or the publication lawful.
More examples and exact wording
| Use of Sora | Decision | Exact wording |
|---|---|---|
| Real product image animated into synthetic action | Disclosure recommended | Product image animated with AI; movement is simulated. |
| Clearly fantastical animation | No disclosure identified | No legal wording identified; use the destination platform's label if required. |
Tool capability, EU law and platform policy are separate evidence layers. This page uses first-party product documentation and primary regulatory sources; it is not legal advice.