IUVM: Iran-linked influence content network
OpenAI ● Covered by 2 sources
OpenAI banned accounts tied to an Iranian influence operation called IUVM that used ChatGPT to churn out propaganda.
Based on reporting by OpenAI — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
OpenAI's latest threat report drops another familiar name into the mix: IUVM, an Iran-linked network that's been peddling pro-Tehran, anti-Israel and anti-US content across the web for years. This time, though, the group wasn't just writing by hand. It leaned on ChatGPT to generate and translate articles, apparently to speed up production and make the material read more naturally in English and other languages.
OpenAI says it caught the accounts and shut them down as part of its ongoing effort to track state-linked influence operations abusing its tools. IUVM isn't a new player in the disinformation game, it's been flagged by researchers going back years for running a web of fake news sites designed to look like independent outlets while actually pushing Iranian government talking points.
What's notable here isn't the message, which is the same stuff IUVM has pushed forever, but the machinery behind it. Translation used to be a real bottleneck for operations trying to reach audiences outside their home language. AI chatbots erase that bottleneck almost entirely, letting a small team produce content that reads like it was written by native speakers in a dozen countries at once.
OpenAI frames this as a win, another operation caught and banned before it could do much damage. And that's fair enough. But the report also quietly confirms what a lot of people suspected: propaganda shops from Iran to Russia to China are treating generative AI as standard-issue equipment now, not a novelty.
My take — AI-written commentary, not fact-checked reporting
I don't think banning a batch of accounts is a solution, it's whack-a-mole with better PR. These operations will keep finding cheaper tools and less scrupulous providers, and the real story is that AI has made propaganda a volume business. Detection and bans matter, but they're a patch, not a fix, for a problem that scales faster than any single company's trust and safety team.
Read more about this at: OpenAI
