AI-generated misinformation surges after China disasters, complicating emergency response
The false content has included AI-generated images of bodies floating in floodwaters, fabricated reports of severe flooding in areas that were not affected, and claims of widespread power outages that prompted some residents to buy emergency supplies
China is facing a rise in AI-generated misinformation following natural disasters, with fabricated images, videos and text spreading rapidly on social media after events such as Typhoon Noul, creating challenges for emergency responders and authorities seeking to provide accurate information.
The false content has included AI-generated images of bodies floating in floodwaters, fabricated reports of severe flooding in areas that were not affected, and claims of widespread power outages that prompted some residents to buy emergency supplies, says the BBC.
Other posts have exaggerated real events. After snakes escaped from a breeding farm during flooding, AI-generated images falsely depicted crocodiles being released into nearby rivers, amplifying public concern.
Keep updated, follow The Business Standard's Google news channel
Experts say the growing availability of artificial intelligence tools has lowered the technical barriers to creating convincing fake text, audio and video, allowing users to generate realistic content with simple prompts.
They say the material is often created to attract followers and online traffic for financial gain through sensational content. Others may share fabricated posts for entertainment without considering the consequences, while some seek to stir public unrest.
Authorities face increasing difficulty in identifying and removing AI-generated misinformation as the technology used to create it develops more quickly than detection systems.
Experts describe the situation as a continuing contest between developers of AI-generated content and those attempting to detect it. They also note that proposals to watermark AI-generated material have limitations because such labels can be removed or altered.
China's legal framework has also struggled to keep pace with advances in artificial intelligence, creating additional challenges for regulators.
The Cyberspace Administration of China has launched a nationwide campaign targeting fabricated information, malicious editing of content and online impersonation. Authorities have established specialised teams to monitor social media platforms, and penalties for creating or distributing false information can include fines and prison sentences of up to seven years.
Researchers say efforts to curb misinformation are complicated by what they describe as a lack of public trust in official information.
They note that authorities have at times downplayed the severity of disasters or reported lower casualty figures than expected, making some members of the public more likely to believe unverified information circulating online.
Experts say improving access to reliable information will be critical to limiting the impact of AI-generated misinformation. They have called for the development of trusted public information systems, including official early warning services and disaster information platforms.
Some also argue that developers of AI applications and social media platforms should face greater legal responsibility for enabling the spread of fabricated content.
Researchers warn that the use of AI-generated misinformation is likely to extend beyond natural disasters into other areas, including political elections and public health emergencies.
