OpenAI announced that it had shut down a series of accounts tied to Chinese influence operations that allegedly used ChatGPT to generate content aimed at shaping American discussions surrounding tariffs and the rapid expansion of AI data centers.

According to a report by Axios, the campaigns failed to gain meaningful attention online, but OpenAI said they offer a glimpse into how individuals connected to Beijing are testing artificial intelligence as a tool for amplifying divisions and contentious policy debates within the United States.

The company disclosed that investigators uncovered two separate efforts that relied on ChatGPT to create social media content, including posts, comments, and political cartoons focused on U.S. technology issues. One campaign, which OpenAI dubbed “Data Center Bandwagon,” produced material arguing that AI data centers were driving up electricity prices for American consumers. A second campaign, labeled “Tech and Tariffs,” generated content attacking the Trump administration’s tariff policies and criticizing America’s efforts to maintain leadership in advanced technology sectors.

“This was not a case of an influence operation creating a debate,” said Ben Nimmo, principal investigator on OpenAI’s intelligence and investigations team, in comments to reporters. “The debate existed already. This was an influence operation from China trying to interfere in it.”

While OpenAI said neither effort achieved significant reach or engagement, company officials noted that the data center operation appears to be the first known China-linked influence campaign to use OpenAI’s technology to inject itself into the ongoing public discussion over AI infrastructure and energy consumption.

Investigators believe the data center effort involved individuals connected to a Chinese government contractor. According to OpenAI, those users instructed ChatGPT to create comic strips and other content focused on power-grid capacity and rising electricity costs. The material was later distributed through what appeared to be fake accounts on X, often accompanied by links to legitimate news stories discussing the energy demands of large-scale data centers.

A separate operation, whose organizers could not be definitively identified, used ChatGPT to produce political cartoons critical of U.S. trade and technology policies. One image cited by OpenAI depicted President Trump wearing pants emblazoned with the American flag and the words “America First,” while holding a hammer marked “Tech Dominance” and striking a barrier labeled “Global Future.”

OpenAI said the campaigns illustrate how foreign actors may increasingly turn to artificial intelligence to quickly generate large volumes of content designed to influence debates over divisive political and economic issues. The company said its ability to detect and remove the accounts underscores both the growing sophistication of foreign information operations and the challenges facing AI companies seeking to prevent their platforms from being exploited for propaganda and influence campaigns.

{Matzav.com}