PRC-linked Influence Operations Target US AI Debates

A recent investigation has uncovered two clusters of ChatGPT accounts linked to China that were involved in covert influence operations aimed at manipulating public opinion on American artificial intelligence policies. The operation, which was detected and disrupted by the platform's moderators, demonstrates a new tactic used by foreign actors seeking to exploit existing concerns about AI infrastructure.

The first cluster, dubbed the 'Data Center Bandwagon' campaign, generated social media comments and images claiming that data center buildouts for AI were driving up electricity prices for average families. This narrative was designed to tap into public worries about energy costs and local impacts of large-scale development projects.

A second cluster, known as the 'Tech and Tariffs' campaign, focused on criticizing US tariffs as attempts to dominate technological competition. Notably, this operation specified that its content should not mention China's leader Xi Jinping, instead featuring President Trump in its output. This suggests a deliberate attempt to avoid drawing attention to Chinese involvement.

The Tech and Tariffs cluster was also linked to a network of likely inauthentic social media accounts targeting OpenAI by spreading false claims about compromised user data. These allegations were entirely unfounded, but they demonstrate the extent to which foreign influence operators will go to create confusion and undermine trust in AI systems.

While the operation appears not to have shifted public opinion significantly, its significance lies in showing that PRC-origin influence operators are testing narratives against US AI infrastructure – a critical component of American technological leadership and economic growth. By exploiting existing concerns about energy prices and data center development, these actors sought to insert themselves into an ongoing debate about the future of US AI capabilities.

The findings highlight the need for greater vigilance in identifying and disrupting foreign influence operations that seek to manipulate legitimate public debates. By publishing this information, the platform aims to help its users, governments, civil society, and the broader public better recognize and counter these threats.