OpenAI has banned clusters of ChatGPT accounts linked to Russian and Iranian influence operations accused of using artificial intelligence to spread geopolitical messaging through fake media identities and research platforms.
The company disclosed the action in a report published on Thursday, saying the two operations used its models alongside conventional techniques to support deceptive campaigns targeting audiences in different countries.
According to OpenAI, the Russian-linked operation targeted audiences across Latin America through a purported research organisation called the Social Research Center (SRC).
The operation allegedly sought to undermine Ukraine’s reputation in the region and influence political developments in countries including Argentina and Bolivia.
OpenAI said the operators used ChatGPT to produce internal reports, create campaign content and prepare documents about the organisation’s activities.
They also generated fake leaked documents and audio scripts, some of which the company identified as having circulated online.
The company said the operation appeared to have involved people in Latin America who were unaware they were working for a Russian group.
OpenAI described the campaign as the most complex attempt to operate a false-front entity it had disrupted in more than two years of reporting on covert influence operations.
The Iranian-linked operation, which OpenAI named “Bogus Bylines”, allegedly used seven fabricated journalist identities to pitch articles to media outlets around the world.
The accounts generated content in Persian and English, including long-form articles focused largely on the conflict involving the United States and Iran.
They also produced social media comments and editorial pitches intended to help place the articles in online publications.
OpenAI said it identified nearly 100 articles published or syndicated under the operation’s bylines across roughly a dozen online outlets. The earliest identified article dated to July 2025, while the latest appeared in October 2026.
Some of the fabricated journalist identities had accompanying social media accounts designed to make them appear legitimate.
OpenAI said transparency information on some accounts indicated links to Iran despite claims that the profiles belonged to Western writers.
The company said it had shared information about the operations with relevant authorities and industry partners.