OpenAI has disclosed that China-linked operatives used ChatGPT to run coordinated influence campaigns targeting American debates over artificial intelligence infrastructure and trade policy, marking a significant example of how generative AI tools can be weaponized for foreign interference. The company banned two clusters of accounts on June 10, 2026, after discovering they had generated thousands of social media comments and images designed to manipulate public opinion on two hot-button issues facing the United States.

What Were the Two Influence Campaigns?

OpenAI identified two distinct operations, each targeting different aspects of American technology and economic policy. The first campaign, dubbed "Data Centre Bandwagon," focused on amplifying opposition to AI data centers by claiming they were driving up electricity costs for ordinary Americans. The second, called "Tech and Tariffs," generated content criticizing the Trump administration's tariffs and framing them as an attempt to dominate technological competition with China.

In the data center campaign, operatives used ChatGPT to create English-language comments and images that blamed AI infrastructure for rising electrical bills. OpenAI released examples showing comic strips depicting a cigar-chomping businessman holding bags of money while a family reacted in shock to their electricity bill, a visual designed to resonate with existing public anxieties about energy costs.

The tariff campaign took a different approach. Operatives generated political cartoons and comments criticizing US trade policy, with explicit instructions to mention only President Donald Trump and avoid any reference to Chinese President Xi Jinping. This selective framing was designed to shift blame for trade tensions entirely onto American leadership.

How Did OpenAI Discover the Campaign?

OpenAI's investigation revealed several telltale signs that exposed the coordinated nature of these operations. The accounts were connected to inauthentic social media profiles, and the prompts given to ChatGPT contained specific instructions that betrayed their foreign origin. Most notably, operatives instructed the AI system to exclude mentions of Xi Jinping while focusing criticism on Trump, a detail that would be unlikely to emerge organically from American users.

The accounts also spread false claims that ChatGPT user data had been compromised, a claim OpenAI explicitly denied. This secondary narrative appeared designed to undermine trust in OpenAI itself while the primary campaigns attempted to influence policy debates.

What Impact Did These Campaigns Have?

Despite the sophistication of the operation, OpenAI found that the campaigns had minimal real-world impact. The company stated that the accounts received "no authentic engagement" and showed "no evidence of meaningful breakout beyond its own activity." This suggests that while the operatives successfully generated content using ChatGPT, they failed to gain traction with genuine American audiences.

However, experts caution that the lack of immediate impact does not diminish the significance of the attempt. Darren Linvill, a professor at Clemson University who studies foreign influence campaigns, acknowledged that Chinese AI-driven influence operations are improving over time. "My team is very familiar with the work of various Chinese influence actors, and the AI work China has done to date has been interesting but not effective," Linvill explained. "It's getting better with each passing month, and I'm concerned what they may be capable of in the future, but they aren't there yet".

Why Is This Significant for US-China Competition?

The disclosure comes amid intensifying competition between the United States and China over artificial intelligence dominance. Both nations are investing heavily in AI infrastructure, advanced computing, and next-generation technologies viewed as critical to economic growth and national security. The use of ChatGPT itself to run these campaigns highlights an irony that OpenAI pointed out: the operatives relied on American AI technology rather than Chinese models to generate their influence content.

Congressional leaders responded swiftly to the disclosure. Rep. John Moolenaar, chairman of the House Select Committee on the Chinese Communist Party, stated that Beijing was attempting to exploit America's open political system. "There are legitimate questions about data centres, and Americans deserve answers as companies work to build the infrastructure we need for our nation's future," Moolenaar said. "Unfortunately, the Chinese Communist Party exploits our openness and works to divide Americans through its United Front organisations and other entities".

How to Identify AI-Driven Influence Operations?

OpenAI's disclosure provides valuable lessons for identifying coordinated inauthentic behavior powered by generative AI. Here are key indicators that researchers and platforms should monitor:

• Suspicious Account Patterns: Accounts that suddenly appear with coordinated messaging, lack authentic engagement history, or show no genuine social connections to their claimed communities warrant closer scrutiny.
• Unnatural Prompt Instructions: Content that includes specific directives to exclude or include particular political figures, or that avoids mentioning certain topics, suggests AI-generated material rather than organic user commentary.
• Rapid Content Generation: Unusually high volumes of similar content posted in short timeframes, particularly images and comments on the same topic, can indicate automated generation rather than authentic user activity.
• Targeting of Existing Divisions: Influence campaigns typically amplify legitimate public concerns rather than inventing new ones, so content that latches onto real debates about energy costs or trade policy should be examined for coordinated origins.
• Cross-Platform Coordination: When the same narratives appear simultaneously across multiple social media platforms through different accounts, it suggests centralized direction rather than organic discussion.

What Does This Mean for Data Center Debates?

Opposition to AI data center construction has been rising in the United States, with at least 36 projects blocked or delayed between May 2024 and June 2025, according to Data Center Watch, a research project by AI security company 10a Labs. The campaigns identified by OpenAI attempted to capitalize on and amplify these genuine public concerns about energy consumption and rising electricity costs.

Data centers supporting AI models like ChatGPT consume enormous amounts of electricity. According to the International Energy Agency, data centers accounted for 1.5 percent of global electricity use in 2024, with consumption growing 12 percent annually over the last five years. These real environmental and economic concerns provide fertile ground for influence operations, which is precisely why foreign actors targeted this debate.

In March 2026, Senator Bernie Sanders and House Representative Alexandria Ocasio-Cortez announced legislation that would impose a moratorium on new data centers until the introduction of national safeguards to mitigate the risks of AI. However, the legislation has little chance of becoming law in the near future due to President Trump's laissez-faire approach to AI regulation and Republican control of both chambers of Congress.

China's embassy in Washington responded to OpenAI's disclosure by stating it was unfamiliar with the report but opposed "any groundless attacks or smears against China." A spokesperson added that "AI is profoundly changing the way people work and live. It is a new frontier for all humanity," and that "China believes in a people-centered approach to AI and advocates openness and inclusiveness to ensure AI is a force for good and for all".

OpenAI's decision to publicly disclose these operations reflects a broader effort to help governments, the AI community, and the general public understand how generative AI can be weaponized by foreign actors. The company emphasized that such operations aim to "manipulate legitimate public debates, weaken democratic institutions and advance totalitarianism with AI characteristics, the use of AI for surveillance, censorship and control over political, social and private life".