Chinese influence operation uses ChatGPT to manipulate Latin American public opinion
OpenAI reported that a Chinese influence operation used ChatGPT to generate anti-U.S. propaganda in Spanish and publish it in Latin American newspapers, as part of a broader campaign to manipulate regional public opinion and counterbalance U.S. influence.
What happened
A Chinese influence operation dubbed 'Sponsored Discontent' used ChatGPT accounts to generate Spanish-language articles critical of the United States, which were published in mainstream newspapers across Mexico, Peru and Ecuador. The articles focused on U.S. political discord and drug addiction, homelessness, racism and crime, with some appearing as sponsored content in Latin American publications.
The operation used ChatGPT to translate and expand Chinese content into Spanish, and produce English-language comments attacking Chinese dissident Cai Xia. Content was created during business hours in mainland China, with some articles bearing the byline of Jil Yousen Culture Communication Co.
Reported impact
The incident highlights the potential for AI to amplify disinformation and accelerate nation-state-backed influence operations targeting regional public opinion.
Index analysis
The operation is part of China's broader strategy to expand its leverage in Latin America through economic entanglements, critical infrastructure investments, and information control. By manipulating public opinion through seemingly legitimate news sources, China aimed to reshape regional alignments and counterbalance U.S. influence.
Classification
Sources and evidence
Sources not yet published
This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.