Chinese Police Use AI-Generated Smear Campaigns to Silences Critics and Opponents

In a chilling display of cyber warfare, Chinese police have been found using OpenAI's ChatGPT AI chatbot to plan and execute smear campaigns against critics and opponents of the Chinese Communist Party. According to OpenAI's latest report on malicious uses of its models, a user with links to Chinese law enforcement attempted to use ChatGPT to run targeted operations against Japanese politicians and dissidents.

The Malicious Use of ChatGPT

In mid-October 2025, the user created a draft plan for a smear campaign against Sanae Takaichi, Japan's first female prime minister. The plan included posting negative content about her on social media, using fake email accounts to send complaints to other Japanese politicians, and launching a series of hashtags targeting Japanese influencers. The malicious activity was designed to discredit Takaichi and silence critics of the Chinese Communist Party.

The user's broader activity across "cyber special operations" moved beyond posting on social media and into exerting social and psychological pressure to silence critics. They targeted dissidents' mental health, hacked their livestreams, and reported their social media accounts for phony violations, sometimes supported by fake evidence.

The Role of ChatGPT in the Smear Campaign

ChatGPT refused to carry out the original smear campaign against Takaichi, but the user later asked it to edit status reports on "cyber special operations." The malicious activity spotted by OpenAI included generating status reports on operations targeting Chinese dissidents and CCP critics, including the specific covert op against Takaichi. These reports claimed the operation asked unnamed Japanese influencers for support and launched a set of hashtags.

The report notes that while some of these posts containing the hashtag alongside Takaichi memes gained little traction, they were still found on social media platforms such as X, Pixiv, and Blogspot. The YouTube videos associated with this campaign had single-digit views, indicating the limited success of the smear operation.

A Broader Pattern of Transnational Repression

OpenAI's report reveals a broader pattern of transnational repression carried out by Chinese authorities using AI-generated content. These operations target people who dare to criticize the CCP's record, not just at home but anywhere in the world.

"This is what Chinese, modern, transnational repression looks like," said Ben Nimmo, principal investigator on OpenAI's Intelligence and Investigations team. "It's not just digital. It's not just about trolling. It's about trying to hit critics of the CCP with everything everywhere, all at once."

These cyber operations are well-sourced and meticulously planned, targeting people who speak out against human rights abuses in China.

Conclusion

The use of AI-generated content by Chinese police to silence critics and opponents is a concerning trend that highlights the dangers of transnational repression. OpenAI's report serves as a warning to anyone using its models for malicious purposes, while also emphasizing the need for greater awareness and vigilance in the face of these types of threats.

As Nimmo noted, "This is not just a digital issue. It's about trying to silence critics of human rights abuses." The use of AI-generated content to carry out smear campaigns against critics of the Chinese Communist Party is a stark reminder of the need for greater security and vigilance in the face of transnational threats.

Keywords: ChatGPT, OpenAI, Chinese police, cyber warfare, smear campaign, transnational repression, human rights abuses.