OpenAI Reports PRC-Linked Influence Operations Targeting U.S. AI Debates

OpenAI says it identified and banned two clusters of ChatGPT accounts that it believes were linked to influence operations originating in China. The company says the activity was aimed at shaping public discussion around U.S. AI policy and broader technology issues. According to the report, one campaign generated comments and images arguing that AI data center construction was raising electricity prices for ordinary families. A second campaign produced content attacking U.S. tariffs and directing outputs to avoid references to Xi Jinping while focusing on President Trump. OpenAI said the activity appeared to be part of covert efforts to amplify existing concerns about energy costs, local development, and U.S. technological competition. The company said it found no evidence that the campaigns spread widely beyond their own networks, but described the effort as an example of foreign actors testing narratives around AI infrastructure and democratic debate.