AI Chatbot Logs Reveal Systematic Campaign Targeting Overseas Critics

A recent investigation by OpenAI has uncovered a covert operation in which Chinese government officials used the ChatGPT platform to draft threats, compile surveillance reports, and create propaganda aimed at Chinese dissidents living outside the country. The findings, first detailed in a company threat report, show how commercially available AI tools are being adapted for state-led intimidation that crosses international borders.

According to the report, officials used the chatbot to compose polished messages in English and other languages to pressure critics of the Chinese Communist Party. The AI was also tasked with summarizing the social media activity of specific dissidents, effectively acting as a monitoring tool. This operation aligns with documented patterns of transnational repression by Beijing, but introduces a new, efficient technological method.

Previously, such campaigns relied heavily on human operatives for surveillance and harassment. The adoption of AI allows a smaller team to generate high volumes of targeted material in multiple languages, changing the operational scale. OpenAI stated it has disabled the accounts involved. Similar state-linked misuse from other countries has been reported before, but this case was notable for its focus on specific individuals rather than general disinformation.

The disclosure arrives as many governments debate controls on advanced AI. The incident highlights the challenges facing technology companies, which must police platform misuse while their tools remain globally accessible. For dissidents and human rights groups, the report confirms long-held suspicions about the sources of threatening communications, but also signals an escalation in how technology can be wielded against them.

Source: Webpronews