OpenAI Exposes Russian and Iranian Influence Operations That Planted Fake News in Real Outlets
OpenAI has banned accounts tied to a Russian disinformation campaign in Latin America and an Iranian operation that placed nearly 100 fabricated articles in online news outlets worldwide.
OpenAI has uncovered and shut down two separate influence operations — one Russian, one Iranian — that used its artificial intelligence tools to spread disinformation, including planting fabricated stories in real news outlets. The company banned the accounts involved, according to its latest threat report.
The Russian campaign, dubbed «Dark Clark», targeted audiences across Latin America. It generated disinformation that was spread online and drew reactions from politicians in the region. OpenAI assigned the operation a category 5 rating out of 6 — the first time any campaign has received such a high severity score in the company's reporting history. The rating reflects the operation's reach and its apparent success in provoking political responses.
The Iranian operation, called «Bogus Bylines», took a different approach. It created seven fictitious journalists and used them to place nearly 100 articles in online outlets around the world. The articles were presented as legitimate news stories, but were in fact fabricated content produced with the help of AI. By mimicking the style and format of real journalism, the operation sought to lend credibility to its narratives and reach unsuspecting readers.
Both campaigns illustrate how generative AI is lowering the barriers to large-scale disinformation. What once required teams of writers and significant resources can now be executed with a handful of accounts and automated tools. The placement of fake stories in real publications is particularly concerning because it exploits the trust readers place in established media brands.
OpenAI's disclosure adds to a growing body of evidence that state-linked actors are experimenting with AI to influence public opinion abroad. The company said it has banned the accounts associated with both operations and continues to monitor for similar activity. The report did not specify which outlets published the Iranian articles or which politicians reacted to the Russian campaign.
The findings underscore the challenge facing platforms and news organizations as they try to detect and counter AI-generated propaganda. While OpenAI's action removes the accounts from its systems, the content they produced may still circulate online, and the tactics they pioneered are likely to be copied by others.
8
