China Faces New Challenges Amid Natural Disasters: The Threat of AI-Generated Fake Videos
Image source, Chinese state media
Chinese officials, already battling the impact of Typhoon Maytas, are now facing a serious new challenge: highly deceptive and dangerous videos created using artificial intelligence (AI).
In recent months, powerful storms and floods have been accompanied on social media by viral fake videos showing flooded areas with swimming corpses, scenes of severe flooding in unaffected regions, and fabricated reports of emergency rescue and relief efforts.
Some videos falsely claim power outages, prompting people to rush to buy supplies unnecessarily. These videos spread misleading information based on fabricated risks.
Early July, following flooding in Jiangzhou city in Guangxi Province, hundreds of snakes escaped breeding grounds, and a fake image purporting to show otters released into a river was circulated.
Authorities try to avoid complications during these high-pressure rescue operations. However, China’s heavy investment in AI technology also risks equipping the public with easy tools to spread deception.
“Many content creators sensationalize disaster news to gain followers because such content attracts more likes and views. They deliberately organize misinformation to attract attention,” said Hu Wang Zhihua, deputy head of online public security at Zhejiang Police Department, in an interview with Chinese government media.
Recently, individuals creating false reports and videos have been arrested, with punishments ranging from imprisonment to fines.
However, experts warn that the crisis of misinformation may persist.
Why do people do this?
Image source, Social media
Professor Chen Bin, writing in the Chinese Communist Party’s Central Party School journal, explained, “AI technology has lowered barriers to creating false information. Now people can produce believable words, images, audio, and videos without complex technical skills. Even if realistic, this content is entirely fabricated and can be produced in large volumes.”
Professor Zhao Fuping from East China University of Political Science and Law said the motivations behind such fake content vary. Some create it for entertainment, others for personal gain through views or clicks, and some deliberately aim to spread terror or social unrest.
“Some users believe the online world is uncontrolled or that their identities cannot be traced,” he added.
A ‘Cat-and-Mouse Game’
Calls have been made to require clear labeling of all AI-generated content. However, as others can remove these labels and edit the content, this may not be a complete solution.
Professor Su Xiaoke from Beijing Normal University advocates for stronger efforts in geographic verification and original source validation for images.
He added, “But AI content detection technology still lags behind content creation technology. We are engaged in a continuous ‘cat-and-mouse game.’ So don’t assume technology alone can easily identify AI content.”
Image source, Social Media
China is making extensive use of AI technology in its ambition to become a leading global technological power.
Since 2017, China has declared AI as a primary driver of national development.
With significant investments flowing into AI companies, competition domestically has intensified. According to Xinhua, there are over 6,000 AI companies in China.
Professor Su Xiaoke emphasizes the need to build reliable information infrastructure in China, including government-operated early warning systems and disaster-related information platforms accessible by news organizations.
The Risk of Growing Distrust in the Government
However, there have been instances of government-controlled media downplaying the severity of disasters, causing public distrust.
In early July, journalists were instructed to use less severe terms such as “opened” or “cracks” rather than “burst” for damage suffered at Liulan Dam near Nanning.
Though possibly an attempt to minimize the damage, the dam failure led to at least 26 deaths caused by floodwaters.
This is not the only incident.
In 2021, local officials in Henan Province were arrested on accusations of concealing or downplaying the deaths of 139 people, causing a significant breach of public trust.
This erosion of trust has made it easier for individuals with malicious intent to spread false and misleading content.
In emergency situations, law enforcement urges the public to rely only on official and verified sources.
Zhao Kai of Ningbo Public Security Bureau stated, “People should be skeptical of disaster information from unknown sources and trust only official information.”
He assured that those spreading false information will face legal consequences, stressing that the internet is not lawless. Public security agencies will continue to rigorously crack down on illegal online activities.
Special task forces have been set up to closely monitor social media platforms, especially short-video platforms, to conduct investigations and enforce strict action.
What Solutions Are Possible?
On July 23, China’s Cyber Administration launched a nationwide campaign against fake content related to floods and other disasters. It directed local offices to identify and punish fabricated statistics, malicious edits, artificial visuals, content that misrepresents historical events, and fake personas impersonating government officials. Social media platforms were urged to tighten content review processes.
According to Professor Zhao, the approach must combine raising public awareness with strict legal enforcement. He believes publicizing cases involving punishments is necessary.
He added, “The impact of AI-generated fake content ultimately depends on how people use the technology, not the technology itself.”
In his view, existing laws are adequate to address the problem.
However, not everyone agrees. Professor Chen warns that AI technology is advancing faster than legal frameworks, raising new legal questions and expanding the misinformation challenge.
Lawyer Peter Pang, working in Shanghai and Washington, notes the global scale of such issues is growing rapidly.
He believes that companies developing AI applications and platforms that publish fake videos should also be held legally accountable.
“The primary responsibility lies with the platforms that allow this content to be published, as they help amplify its reach,” he told the media.
Professor Su issued a warning: “AI-generated misinformation will not only increase in volume but can spread at much faster rates.”
“In the future, political elections, public health crises, armed conflicts, natural disasters, and issues critically affecting people’s lives may face heightened risks from AI-created false information.
Instead of reacting only after misinformation goes viral, governments and platforms must begin proactive preparations.”
Additional information: BBC News Nepalese content is also available on YouTube. Subscribe to our channel and watch our published videos here. You can also follow us on Facebook, Instagram, and Twitter. Additionally, BBC Nepali Service programs can be heard on Rang Radio every evening at 8:45 PM from Monday to Friday.