OpenAI Bans Iran-Linked ChatGPT Op That Placed ~100 Articles

OpenAI said a cluster of ChatGPT accounts originating in Iran used seven fake journalist personas to place nearly 100 opinion articles across more than a dozen online publications, the Washington Post reported. The company called the campaign "Bogus Bylines" and said activity began in July 2025, ramping up as the US-Iran war escalated.

The personas "presented themselves as Western journalists, mostly Americans," writing about international politics, human rights and the Middle East. In its own disclosure, OpenAI said operators used ChatGPT to edit articles, tailor them to publications' submission requirements and draft emails pitching the stories to editors. Sample pieces carried titles including "Is America's Iran Policy Being Privatized? The Jared Kushner Question" and "The Iran War and the Collapse of Western Moral Credibility."

OpenAI called the activity "consistent with a commercial, hire-for-service influence operation" and said it "could not establish the identity of the specific actor responsible," stopping short of attributing the campaign to the Iranian government. It rated the article-placement effort as Category 4 on OpenAI's six-point Breakout Scale, below a Russian operation disclosed the same day that the company called its first Category 5 case since it began reporting.

Beyond the opinion pieces, the operators generated batches of English and Persian social media comments, including hostile Persian-language replies to posts by Iran International. OpenAI said it has disrupted 30 covert influence operations since early 2024.