Fake AI Videos Are Spreading Fear During China's Disasters

When Typhoon Noul hit China over the weekend, rescue teams were not the only ones racing to respond. Fake videos made with Artificial Intelligence (AI) were already spreading across social media. These videos showed false images of flooding, dead bodies, and panicked crowds. Chinese officials are now worried that AI tools are making it too easy for people to spread dangerous lies during disasters.

Over the past few months, powerful storms and floods have struck several parts of China. Each time, fake videos flooded social media right along with the water. Some clips falsely showed bodies in floodwaters that never happened. Others claimed towns were badly flooded when they were completely dry.

In some cities, fake videos about power outages caused people to rush out and panic buy supplies. People were scared by something that was not even true. This kind of panic is dangerous, especially when real emergencies are happening nearby. Officials say these false posts put real lives at risk.

One striking example happened in Hengzhou city in July. Real flooding caused hundreds of snakes to escape from a snake farm. Then fake images appeared online showing crocodiles being released into local rivers. No crocodiles were ever set free, but many people believed the story because it seemed connected to a real event.

Experts warn that AI is making this problem much worse. In the past, creating a convincing fake video took special skills and expensive tools. Now, anyone can type a simple request into an AI program and get a realistic-looking video in minutes. Professor Chen Bin wrote that AI 'dramatically lowers the technical barriers to producing false information.'

Professor Gao Fuping spoke to the BBC about why people share fake disaster content. He said some do it just for laughs, without thinking about the harm. Others want to gain followers or earn money from online views. And some do it on purpose to cause fear and social unrest.

Police official Huang Zhihua explained that creators know scary stories get more clicks. 'Many of these content creators use disaster news to try to gain followers, knowing that more sensational or exaggerated content will garner more likes and more views,' he said. He added that they are 'deliberately fabricating information to grab people's attention.' This means they are making up lies just to get famous online.

One idea to stop fake content is to put a clear label on all AI-made videos. But experts say people can simply re-edit the video to remove the label. Professor Xu Xiaoke said better tools are needed to check where images come from. However, he warned that technology for spotting fakes still falls behind the technology used to make them.

China has invested billions of dollars in AI and has more than 6,000 AI companies today. This growth has made AI tools widely available to the public. That is good for business, but it also makes it easier to misuse the technology. The same tools that power useful apps can also be used to create harmful fake content.

Police are asking people to be skeptical of disaster news from unknown sources and to trust official announcements instead. Officers have set up a special task force to watch social media closely. People who create fake disaster content can be fined or jailed for up to seven years in serious cases. On July 23, China's top internet regulator also launched a nationwide crackdown on fake online flood content.

However, trusting official sources is also complicated. Earlier this month, officials told reporters to call a broken dam wall an 'opening' instead of a 'breach.' The damage was far worse than officials admitted, and at least 26 people died. In 2021, officials were arrested for hiding 139 deaths during floods in Henan province.

These cover-ups have made many people unsure of who to believe. When trust in the government is low, fake content spreads more easily. Professor Gao believes the best answer is public education combined with stronger laws. He said people need to see real examples of those who were punished for spreading fake content.

Lawyer Peter Pang says courts around the world are seeing more and more cases about AI misinformation. He believes both video creators and the platforms that host fake content should be held responsible. 'The first line would be the platforms, because they're the ones that are the enablers,' he told the BBC. He argued that if a fake video causes panic and someone gets hurt, the video's creator shares the blame.

Experts agree that AI fake news will keep growing in the future. Professor Xu warned it will increase not just in amount, but in how fast it spreads. He said elections, health emergencies, wars, and natural disasters are all high-risk areas. His message is clear: governments and platforms must prepare now, before the next wave of fake content hits.