By Gloria Ogbonna
OpenAI announced Wednesday that it has banned a number of accounts linked to Chinese influence operations that allegedly used ChatGPT to generate content designed to shape online discussions in the United States about tariffs, artificial intelligence, and the growing debate surrounding AI data centers.
The disclosure sheds light on how foreign actors are increasingly experimenting with advanced AI tools to amplify political narratives, influence public opinion, and exploit existing divisions within American society. While the campaigns reportedly failed to gain significant traction online, OpenAI officials say the operations demonstrate an evolving strategy in which artificial intelligence is used to mass-produce persuasive content at scale.
According to a report by Axios, OpenAI investigators uncovered two separate influence campaigns that relied on ChatGPT to create social media posts, comments, and political cartoons centered on controversial topics in American technology and economic policy.
The first operation, which OpenAI internally dubbed “Data Center Bandwagon,” focused on the rapidly expanding network of AI data centers across the United States. The campaign attempted to convince Americans that these facilities were contributing to rising electricity prices and placing increasing strain on the nation’s power grid.
The second operation, referred to as “Tech and Tariffs,” concentrated on criticism of U.S. trade policy, particularly tariffs associated with President Donald Trump’s administration, as well as broader American efforts to maintain technological leadership over global competitors, including China.
OpenAI’s intelligence and investigations team concluded that both campaigns used the company’s AI models to generate large volumes of content intended for publication across social media platforms. The content included written commentary, political messaging, and illustrated cartoons designed to attract attention and encourage engagement around divisive policy issues.
Ben Nimmo, a principal investigator on OpenAI’s intelligence and investigations team, emphasized that the campaigns did not create entirely new controversies but instead attempted to exploit debates that were already taking place in American political discourse.
“This was not a case of an influence operation creating a debate,” Nimmo told reporters. “The debate existed already. This was an influence operation from China trying to interfere in it.”
According to OpenAI, neither campaign managed to achieve widespread influence or generate meaningful engagement. Nevertheless, company officials view the incident as an important warning sign about the future use of artificial intelligence in information warfare and political influence efforts.
One OpenAI official told reporters that the operation targeting discussions about AI data centers appears to be the first known instance in which a China-linked influence campaign used OpenAI’s technology specifically to intervene in public debates over data center infrastructure and energy consumption.
Investigators believe the data center-focused campaign was operated by individuals connected to a Chinese government contractor. Those users allegedly instructed ChatGPT to create political cartoons and social media content addressing topics such as electricity shortages, rising energy prices, and power-grid reliability.
The generated materials were later posted on the social media platform X through accounts that OpenAI described as likely inauthentic. Many of the posts included links to legitimate news reports discussing the growing energy demands of AI facilities, an issue that has become increasingly important as major technology companies race to expand artificial intelligence capabilities.
By combining AI-generated commentary with authentic news coverage, the operators appeared to be attempting to lend credibility to their messaging while encouraging users to view AI infrastructure as harmful to everyday American consumers.
A separate campaign focused on trade and technology policy employed similar tactics. OpenAI said users involved in the operation utilized ChatGPT to create political cartoons and commentary critical of American economic and technological ambitions.
One cartoon cited by investigators depicted President Trump wearing pants decorated with the American flag and carrying the slogan “America First.” In the image, Trump was shown using a large mallet labeled “Tech Dominance” to strike a wall marked “Global Future,” symbolizing criticism of U.S. policies aimed at maintaining technological leadership and protecting domestic industries.
While OpenAI was able to identify suspicious activity connected to the campaign, investigators noted that they could not definitively attribute this second operation to a specific organization or government entity.
The company described both campaigns as examples of how foreign influence actors are adapting to rapidly advancing AI technology. Rather than relying solely on human operators to create content, organizations can now use artificial intelligence to generate thousands of posts, comments, images, and political messages in a fraction of the time previously required.
Security experts warn that such capabilities could become increasingly attractive to state-sponsored influence operations seeking to shape public opinion, spread propaganda, or deepen political divisions abroad.
OpenAI said the discovery and removal of the accounts underscores the growing responsibility facing AI companies as their technologies become more powerful and widely accessible. Detecting and preventing the misuse of AI systems has become a major focus for technology firms, governments, and cybersecurity researchers around the world.
The incident also highlights the broader strategic competition between the United States and China in the field of artificial intelligence. As both nations invest heavily in AI development, concerns continue to grow over how the technology could be used not only for economic and scientific advancement but also for information warfare, political influence campaigns, and national security operations.
Experts have increasingly warned that AI-generated content may make it easier for foreign actors to produce convincing propaganda, create misleading narratives, and flood social media platforms with coordinated messaging that appears authentic to ordinary users.
Although the campaigns identified by OpenAI did not achieve significant success, company officials believe they represent an early glimpse into a future where artificial intelligence plays a much larger role in geopolitical competition and digital influence operations.
As governments, technology companies, and researchers work to establish safeguards against misuse, the discovery serves as another reminder that AI is rapidly becoming both a transformative technological tool and a new battleground in the global struggle for influence, information, and technological leadership.
Source: Breitbart
Original article
Comments