OpenAI said it is collaborating with cybersecurity researchers and law enforcement agencies to counter such activities and strengthen detection mechanisms for AI-generated disinformation
OpenAI has uncovered evidence of a Chinese state-linked surveillance tool leveraging artificial intelligence (AI) to generate content and support online information operations. The revelation, made in the company’s latest security report, highlights the growing use of AI to amplify disinformation campaigns and enhance cyber capabilities on a global scale.
The tool, known as Spamouflage, is believed to be operated by the Chinese Communist Party (CCP), according to cybersecurity researchers. It reportedly generates and disseminates large volumes of content across social media platforms, targeting both domestic and international audiences. The system automates the creation of social media posts, comments, and articles in multiple languages to influence public opinion and obscure critical narratives.
OpenAI disclosed that the system used advanced language models to craft highly convincing, human-like messages, making them more difficult to detect. The company identified that the tool had been accessing its AI services to generate content before its access was detected and subsequently blocked.
“Our investigation uncovered the use of our AI models by state-affiliated actors to amplify disinformation and manipulate online conversations. We have taken action to disrupt this activity and will continue working to prevent misuse of our technology,” OpenAI said in a statement.
While the company did not specify the duration of the tool’s operation, cybersecurity experts suggest that it signals a significant advancement in online influence campaigns. Unlike traditional disinformation tactics that relied on repetitive, low-quality content, AI-powered systems produce highly realistic text, making detection more challenging for users and platforms alike.
The discovery highlights mounting concerns about the intersection of AI and cyber warfare. With AI systems becoming more sophisticated, governments and technology companies are under pressure to implement safeguards against the misuse of AI tools for malicious purposes.
China has previously denied allegations of state-sponsored disinformation campaigns. However, several cybersecurity firms have consistently pointed to evidence of coordinated influence operations linked to the country.
OpenAI said it is collaborating with cybersecurity researchers and law enforcement agencies to counter such activities and strengthen detection mechanisms for AI-generated disinformation. The company emphasised its commitment to preventing the misuse of its technology while advancing the development of responsible AI systems.
