OpenAI claims Chinese government uses ChatGPT, and the company is quite unhappy about this as ...

OpenAI has alleged that the Chinese government has been using ChatGPT in its cyber operations. The company says that this highlights the dangers of AI misuse. According to a report by Business Insider, OpenAI’s new report, ‘Disrupting Malicious Uses of AI’, which documents how Beijing’s operatives leveraged the chatbot to polish internal status reports tied to influence campaigns.

As reported by Business Insider, OpenAI said that it uncovered the operation after removing an account linked to the Chinese government which had been uploading periodic reports to ChatGPT for editing. Following this, OpenAI did a deeper investigation into China’s ‘cyber special operations’, which target potential dissidents both inside and outside China and abroad.

Large scale campaigns

The reported further described the efforts as ‘large-scale, resource-intensive and sustained,” inviting hundreds of staff, thousands of fake accounts and dozens of tactics.

OpenAI also revealed that the targets included dissidents worldwide and even foreign leaders such as the Prime Minister of Japan. OpenAI has also documented the cases where Chinese operatives forged the US country court documents in order to create a pressure on social media platforms into removing posts.

The ChatGPT-maker also found out that the coordinated campaigns were fake accounts submitted abusive reports en masse to trigger bans on dissident voices.

Some of these reports even used AI-generated images posing as screenshots of conversations.

Dissidents speak out

As reported by BI, one prominent target has been the X account @whyyoutouzhele, better known as “Teacher Li is not your teacher,” which has more than 2.1 million followers and frequently posts videos exposing corruption and human rights abuses in China. The account’s team warned: “Your content moderation system is being used by the Chinese Communist Party as a weapon.”

They urged the AI industry to take responsibility, adding: “When your technology is being used to systematically oppress human rights, to say that ‘we’re just makers of a tool’ is not an acceptable answer.

”Other platforms have acknowledged similar patterns. Bluesky said it recently removed accounts engaged in coordinated inauthentic activity, while Meta confirmed it tracks such operations in its adversarial reports. X and China’s Ministry of Foreign Affairs did not respond to requests for comment.