Recently, the Pingliang Public Security Bureau’s Network Security Brigade in China cracked a case of using ChatGPT to fabricate fake information.
On April 25th, during a routine online patrol, the cybersecurity team of Kongtong Branch of the local public security bureau discovered an article with the title “Train Collides with Road Workers in Gansu This Morning, Resulting in 9 Deaths” on a Baijiahao (news platform of Baidu) account. It was preliminarily determined that the information was false and untrue.
The cyber security police immediately launched an investigation and found a total of 21 Baidu accounts that had published the article during the same time period. The content of the article mentioned locations including many places. The article has been clicked over 15,000 times.
After investigation, all the Baidu accounts involved in the case belong to a self-media company in Shenzhen. The legal representative of the company, Mr. Hong, is suspected of committing a major crime.
On May 5th, the police officers collected evidence from the computer and Baijiahao used by the suspect at his residence in Dongguan, Guangdong Province.
After interrogation, the suspect Hong learned about a method of earning money from online traffic through WeChat friends and purchased a large number of “Baijiahao” accounts.
In order to avoid the plagiarism detection function of Baijiahao, Hong used the popular ChatGPT artificial intelligence software to modify and edit the collected news elements recently, and illegally uploaded them to his purchased Baijiahao account for profit.
Hong used modern technology to fabricate false information and spread it on the internet, which was widely circulated and viewed. His behavior is suspected of constituting the crime of provocation and troublemaking.
Currently, the public security authorities in the region have taken criminal coercive measures against the suspect Hong and the case is under further investigation.