US EditionOpenAI Uncovers Global Chinese Intimidation Operation Through One Official’s Use of ChatGPT
Google and partners disrupted UNC2814's use of the GRIDTIDE backdoor, which targeted 53 organizations in 42 countries with a decade-long espionage campaign.
- This week, OpenAI released a threat report revealing a single ChatGPT account tied to an unnamed Chinese law‑enforcement official that documented a global influence and intimidation campaign using status reports on 'cyber special operations' targeting Chinese dissidents and CCP critics.
- Beginning in late October 2025, hashtags appeared attacking Japan's prime minister and criticizing US tariffs, with Hui Bo's X account restricted by November 29, 2025, OpenAI said.
- OpenAI's report details tactics such as forged documents, impersonation of U.S. immigration officials, fake obituaries for dissident Jie Lijian, and mass reporting targeting activist Hui Bo, involving hundreds of human operators and thousands of fake accounts.
- After the report, OpenAI banned the user and said investigators matched ChatGPT entries to real-world impacts, while Google Threat Intelligence Group disrupted UNC2814’s infrastructure and notified 53 victims in 42 countries.
- Amid a broader U.S.-China AI rivalry, the report shows threat actors use multiple AI models as amplifiers, highlighting industry tensions and debates over AI's role in defense and influence operations.
29 Articles
29 Articles
OpenAI announced that it has suspended more accounts associated with China's order forces, specialized shells in romantic shelters and influence operations, including a campaign of denial against the first woman...
A vast Chinese influence operation, accidentally revealed by a Chinese law enforcement officer's use of ChatGPT, focused on intimidating Chinese dissidents abroad, including by impersonating US immigration officials, according to…
Google defeated a group of hackers connected to China that invaded at least 53 organizations in 42 countries, according to the company itself, this Thursday. Google and unidentified partners closed projects of...
Google disrupts Chinese-linked hackers that attacked 53 groups globally
Google disrupted a Chinese-linked hacking group that breached at least 53 organizations across 42 countries, the company said Wednesday, February 25. The hacking group, tracked as UNC2814 and “Gallium,” has a nearly decade-long history of penetrating government organizations and telecommunications companies, the company said in findings shared exclusively with Reuters. “This was a vast surveillance apparatus used to spy on people and organizatio…
Coverage Details
Bias Distribution
- 68% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
