Fact Check: Partly AI-Generated Video Misrepresented as Authentic Footage of the 2026 Nepal Floods

On August 26, 2026, a massive flood occurred along the Nepal-China border. Following the disaster, many social media users shared images and videos purportedly related to the flooding. However, some of this content was unrelated to the disaster, shared in a misleading context, or generated using artificial intelligence.

One such example is a video shared on X on August 29, 2026, by a user named Riyaz_H_Khan. The 15-second video purportedly shows scenes from the flood disaster in Nepal. Within one hour of being posted, the video had been viewed more than 1,025 times, reposted seven times, and liked 18 times.

 

However, a subsequent investigation revealed that the video contains both AI-generated and authentic footage. Specifically, the first five seconds of the video were identified as AI-generated, while the remaining footage appears to be authentic. To assess the authenticity of the content, the video was examined using HIVE Moderation and UncovAI.

The video was initially analyzed using HIVE Moderation, which detected indicators of AI-generated content in the footage. Further examination showed that these indicators were associated with the first five seconds of the video.

The footage was subsequently analyzed using UncovAI, which also detected AI-generated content. The findings from both tools therefore indicate that the video combines AI-generated imagery with authentic footage rather than consisting entirely of genuine documentation of the Nepal floods.

In conclusion, the investigation indicates that the 15-second video should not be considered entirely authentic footage of the August 2026 floods in Nepal. Analyses conducted using HIVE Moderation and UncovAI identified the first five seconds of the video as AI-generated. The video therefore combines artificially generated imagery with real footage while being presented to viewers as a continuous recording of the disaster.

This case highlights an important challenge in identifying AI-generated misinformation: manipulated content does not necessarily have to be entirely artificial. AI-generated segments can be combined with authentic footage, making misleading videos more difficult for viewers to recognize. As generative AI becomes increasingly integrated into visual content, verifying individual segments of a video—as well as its overall context—is essential, particularly when assessing footage associated with breaking news and natural disasters.

 

If you suspect that a video, image, or audio file has been created using artificial intelligence or deepfake technology and would like free assistance in verifying its authenticity, you may send the link to the content or the file itself to allaboutdeepfake@gmail.com