OpenAI says it has uncovered and disrupted two China-linked influence operations that used ChatGPT to generate content aimed at shaping public opinion on artificial intelligence, data centers, tariffs, and broader U.S. technology policy.
According to the company, one campaign focused on America’s rapidly expanding AI infrastructure. Dubbed “Data Center Bandwagon,” the operation allegedly generated social media posts and images claiming that new AI data centers were driving up electricity prices for everyday Americans. Whether you support data center expansion or oppose it, OpenAI argues that foreign actors were attempting to insert themselves into a legitimate domestic debate.
A second campaign, called “Tech and Tariffs,” reportedly produced content criticizing U.S. tariffs and portraying them as an effort to dominate technological competition. OpenAI says prompts connected to the operation specifically instructed the model to feature President Trump while avoiding references to Chinese leader Xi Jinping. The company also linked the effort to a network of likely fake social media accounts that pushed false claims that ChatGPT user data had been compromised.
What caught my attention is not that these campaigns appear to have been particularly successful. OpenAI says it found no evidence of meaningful impact beyond the activity generated by the operations themselves. Instead, the bigger story is that AI infrastructure has apparently become important enough to attract foreign influence efforts.
The obvious question is why China would care about American data centers and AI policy in the first place.
The answer may be fairly simple. AI is increasingly viewed as a strategic technology on par with semiconductors, telecommunications, and energy infrastructure. The United States and China are competing to lead the next generation of AI development, and the country with the strongest models, most computing power, and largest supporting infrastructure could gain significant economic and geopolitical advantages.
If foreign influence operators can slow data center construction, amplify public opposition, or increase political pressure against AI investments, even marginally, that could make it harder for the United States to expand the infrastructure needed to support advanced AI systems. Likewise, efforts to shape discussions around tariffs and technology policy could potentially influence public perception of measures designed to limit China’s access to advanced technologies.
That does not mean every criticism of data centers, electricity usage, or AI policy is foreign propaganda. Many of those concerns are legitimate and deserve serious debate. OpenAI itself noted that the campaigns appeared to latch onto existing concerns rather than invent entirely new ones.
Data centers are increasingly being treated as strategic assets. They consume enormous amounts of power, require significant local investment, and sit at the center of America’s AI ambitions. Debates over their environmental impact, energy consumption, and economic benefits are already heated. OpenAI believes the operators behind these campaigns attempted to exploit those existing concerns rather than create new controversies from scratch.
The company says publishing the findings is part of a broader effort to help governments, businesses, researchers, and the public identify attempts by foreign actors to manipulate democratic discussions using AI-generated content. If OpenAI’s findings are accurate, they offer a reminder that America’s AI race with China is no longer limited to chip factories and research labs. Increasingly, the battle may be playing out online, where influencing public opinion can sometimes be cheaper than building technology.
Support independent tech journalism
NERDS.xyz is independently owned and operated. If you enjoy my coverage of Linux, AI, hardware, cybersecurity, and tech culture, consider supporting the site on Ko-fi.
Support NERDS.xyz