OpenAI's latest threat report reveals two coordinated influence operations originating from China that exploited ChatGPT to sway American tech policy discussions. The company identified and removed dozens of accounts involved in what they termed the 'Data Center Bandwagon' and 'Tech and Tariffs' campaigns, both designed to covertly shape narratives around critical infrastructure and trade policy.

These operations focused on amplifying concerns about AI data centers driving up electricity costs for families, while simultaneously criticizing US tariffs on technology imports. Notably, the second cluster specifically instructed prompts to exclude references to China's leader while emphasizing President Trump, suggesting efforts to frame debates in particular political directions. The accounts also spread false claims about ChatGPT user data being compromised.

Foreign actors have repeatedly targeted domestic issues to build credibility and exploit divisions within democratic societies. While OpenAI found no evidence these campaigns achieved significant reach beyond their own networks, the incident demonstrates how authoritarian regimes are actively testing AI-enabled influence tactics against foundational technology infrastructure. China-linked operatives are probing vulnerabilities in the information ecosystem surrounding AI infrastructure, which underpins both economic competitiveness and democratic discourse.

This represents a shift toward weaponizing AI tools themselves in information warfare, moving beyond simple social media manipulation to exploiting the very platforms shaping public understanding of technology. OpenAI's disclosure serves as a warning that democratic institutions face evolving threats from adversaries who understand that controlling the narrative around AI means influencing the future of technological development itself.