Skip to content
Submit incident
AII-1001-41991
ReportedReported by fewer than three independent publishers, with no official finding.Supporting sources are pending verification.

Sora deepfake video of Spanish coup circulates as misinformation

A deepfake video generated with OpenAI's Sora tool falsely depicting a military coup in Spain circulated widely on social media with over 6 million views. Fact-checkers confirmed the video was entirely synthetic and that no coup had occurred.

15 October 2025 Β· OpenAI Β· Misinformation and synthetic media
15 October 2025
AII-1001-41991
Quick facts
Category
Misinformation and synthetic media
Country
Spain
Date
15 October 2025
Drafted by
Raidu Incident Detection

What happened

A video generated using OpenAI's Sora tool depicting an armed military incursion into Spain's Congress circulated on X and TikTok with over 6 million views. The 10-second video showed soldiers pointing weapons at officials and contained audio of an arrest, accompanied by false claims that the Garde civile and Spanish army had entered Congress to arrest corrupt legislators.

Fact-checking revealed the video was artificial. The original video, published on 15 October 2025 by a TikTok account describing itself as a creator of AI image content, bore Sora's watermark. Users subsequently removed or obscured the watermark before sharing the video widely, which accumulated over 254,000 views on the original post.

Reported impact

The synthetic video reached over 6 million views across X and TikTok and was presented as real documentation of a coup attempt, spreading false claims about military action against Spanish government officials.

Index analysis

Sora allows users to generate photorealistic videos with minimal effort. The tool applies a watermark to all generated content, but removal tools exist and users in this case obscured the Sora branding with emojis. The incident illustrates how synthetic video generation can blur the line between authentic and artificial content on social platforms.

Classification

Category
Misinformation and synthetic media
Industry
Media, social platforms and marketing
Country
Spain
Harm status
Potential harm
Intent
Unknown
Timing
Post-deployment
Harm types
Public interest
Harmed
General public
Failure modes
Deepfake video
Consequences
None recorded
AI systems
Sora

Sources and evidence

Sources not yet published

This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.

How sources are verifiedSuggest a source for this incident