OpenAI has observed an uptick in the use of its AI technology by Chinese groups for covert operations. , Reuters reported, citing a new report released by the ChatGPT developer. The company noted that although these operations have expanded in scope and tactics, they generally remain small-scale and target limited audiences.
The US-based AI firm regularly publishes reports on malicious activities detected on its platform, which include creating and debugging malware or generating fake content for websites and social media.
OpenAI bans accounts for sensitive content generation
OpenAI said that it banned accounts that utilised ChatGPT to produce social media content on political and geopolitical topics related to China. The banned accounts engaged in activities such as criticising a Taiwan-centric video game, making false accusations against a Pakistani activist, and discussing issues related to the closure of USAID. Some accounts also generated posts critical of US President Donald Trump’s tariffs.
The company also detected an influence campaign originating from China, designed to produce polarised content on social media. This campaign sought to bolster both sides of divisive political debates in the US, utilising AI to generate both written content and profile images.
The report mentions another example where China-associated threat actors used AI to aid different phases of their cyber activities. This included engaging in open-source research, altering scripts, troubleshooting system setups, and developing tools for password brute-forcing and automating social media processes.
China’s foreign ministry stated there is ‘no basis’ for OpenAI’s claims when Reuters approached it for comment on the issue.
This is not the first time OpenAI has taken action against malicious accounts. Earlier this year, OpenAI removed accounts linked to China and North Korea due to concerns over the misuse of AI technology. Some of these accounts generated AI-written news articles in Spanish criticising the US. The articles were then published by mainstream news outlets in Latin America under the byline of a Chinese company.
In another case, individuals suspected of North Korean ties allegedly used AI to create fake resumes and online profiles. These fraudulent documents were then used in attempts to secure employment at companies based in Western countries. However, OpenAI did not disclose the number of accounts banned or the timeframe of these actions.
With more than 400 million weekly active users, OpenAI’s ChatGPT is one of the most widely used AI chatbots. In April 2025, OpenAI released two new AI models, o3 and o4-mini, to expand ChatGPT capabilities.
Read more: OpenAI restructuring may lead to future IPO, says CFO
