Skip to content
Submit incident
AII-1001-43833
Well documentedThree or more independent publishers report the core facts.Supporting sources are pending verification.

AI-Generated Videos Spread Misinformation During Hurricane Melissa in Jamaica

AI-generated videos created with OpenAI's Sora model and bearing the company's watermarks proliferated on social media, particularly TikTok, as Hurricane Melissa approached Jamaica in October 2025. The fake videos depicted exaggerated scenarios and misrepresented the hurricane's danger, potentially diverting people from critical safety information about the catastrophic storm.

27 October 2025 Β· OpenAI Β· Misinformation and synthetic media
27 October 2025
AII-1001-43833
Quick facts
Category
Misinformation and synthetic media
Country
Jamaica
Date
27 October 2025
Drafted by
Raidu Incident Detection

What happened

Dozens of AI-generated videos bearing OpenAI's Sora watermarks proliferated on social media feeds, including TikTok, as Hurricane Melissa approached Jamaica on October 27, 2025. The fabricated videos depicted dramatic newscasts, severe flooding, sharks in water, and scenes of human suffering, as well as locals with exaggerated Jamaican accents partying, boating, and minimizing the hurricane threat.

Jamaica's Minister of Information and Senator Dana Morris Dixon stated on Monday that she and other ministers held a press conference to provide correct information, noting she was seeing many false videos in WhatsApp groups and urging people to listen to official channels.

Reported impact

The fake videos diverted attention from critical safety information about the Category 5 storm. According to Amy McGovern, a meteorology professor at the University of Oklahoma specializing in AI for extreme weather forecasting, the fake content undermines the government's serious work and would result in loss of life and property.

The false content led users to underestimate the danger of the hurricane, which forecasters warned could be the island's most violent.

Index analysis

Most of the identified fake videos bore OpenAI's Sora watermark. While TikTok was supposed to require AI-generated content to be labeled as such, only some of the videos carried the mandatory label, and TikTok had begun removing some content by Tuesday.

Governance controls

OpenAI did not respond when contacted by AFP about the misinformation. Jamaica's government held a press conference to distribute official safety information and urged citizens to rely on official channels.

Classification

Category
Misinformation and synthetic media
Industry
Media, social platforms and marketing
Country
Jamaica
Harm status
Potential harm
Intent
Unknown
Timing
Post-deployment
Harm types
Public interest
Harmed
General public
Failure modes
Deepfake video
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