AI Disinformation in Action: OpenAI Uncovers Iran-Linked Influence Operation

What Actually Happened

The operatives used ChatGPT accounts — prompting the tool in Farsi — to refine articles and editorial pitches, then submitted that content to real publications under fake bylines.

Some of the article titles were deliberately provocative:

• "Is America’s Iran Policy Being Privatized? The Jared Kushner Question"
• "The Iran War and the Collapse of Western Moral Credibility"

One piece ran in the LA Progressive, a self-described independent publication based in Los Angeles. Another outlet, the London-based Middle East Monitor, has nearly two million Facebook followers — giving the content meaningful reach.

After the articles published, the same operatives asked ChatGPT to generate fake reader comments for social media, framing the US as the aggressor and Iran as a resilient victim.

Why This Matters Beyond the Headlines

OpenAI banned the accounts involved and flagged the activity in a public blog post. But the more important signal here is structural.

This wasn’t a sophisticated state-level cyberattack. It was a relatively low-cost content operation — fake personas, AI-polished writing, and basic media outreach. The barrier to running this kind of campaign is now extremely low.

Jessica Brandt, former head of the Office of the Director of National Intelligence’s Foreign Malign Influence Center, put it plainly: the operatives managed to land content in real outlets, which expanded the campaign’s potential reach significantly. AI didn’t just assist the operation — it made the false fronts more convincing and harder to detect.

OpenAI itself acknowledged uncertainty about the full scope:

The overall activity appeared consistent with a commercial actor running a for-hire influence campaign, but we are unable to identify the particular actor involved.

The episode adds to broader concerns around AI misinformation and how easily synthetic content can move through trusted channels.

The Broader Pattern

This case doesn’t exist in isolation. It’s part of a wider trend of AI being used to shape public perception around active conflicts.

• Iranian operatives used AI-generated Lego-style videos mocking US leaders, which went viral.
• President Trump posted AI-generated videos depicting US strikes on Iran.
• A separate incident this spring involved a false AI-assisted US intelligence report about a Chinese ship that nearly triggered a military boarding.

The common thread: AI is lowering the cost and raising the sophistication of information warfare on all sides.

That makes accurate verification and faster methods for debunking disinformation increasingly important.

What This Means for Media Trust and AI Tool Platforms

For publishers, the takeaway is uncomfortable. If your outlet accepts unsolicited pitches, you’re a potential distribution channel for AI-assisted propaganda. The Middle East Monitor’s director told CNN he would investigate and "take appropriate action" — but the content had already published and circulated.

For AI platforms, this puts pressure on detection and enforcement. OpenAI caught this operation and acted, but the question is how many similar campaigns are running undetected across other tools and platforms.

For anyone consuming news online, the practical implication is straightforward: the byline is no longer a reliable signal of authenticity. A polished article from a named journalist at a real-sounding outlet can now be the output of a coordinated AI-assisted influence campaign.

That also raises questions about long-term verification skills and the role of better verification practices in editorial and publishing workflows.

The Takeaway

AI disinformation isn’t a future problem — it’s a current one with documented, real-world examples. The tools are accessible, the cost is low, and the playbook is simple: create fake personas, use AI to produce credible-sounding content, and find legitimate outlets willing to publish it.

The defense isn’t just platform-level enforcement. It’s editorial skepticism, better verification practices, and a clear-eyed understanding that AI-generated content can be indistinguishable from human-written work when the goal is deception.

Observe the tools. Watch how they’re being used. Choose what you trust carefully.