OpenAI disclosed a troubling case: according to the company, Iranian operatives used ChatGPT to create and publish more than 100 articles criticizing the U.S. war against Iran.

The operation involved seven “fake journalists,” and the content appeared in roughly a dozen media outlets around the world.

What’s really worth paying attention to isn’t that AI can write articles, but that AI is dramatically lowering the cost of producing content for information warfare.

In the past, carrying out a cross-border influence campaign required a whole team of people, including reporters, editors, translators, and operations staff. Now, models can quickly generate, rewrite, and translate content, allowing a small number of people to produce volumes that once required a much larger team.

And the “fake journalist” persona may be even more concerning than the 100-plus articles.

That’s because information warfare doesn’t necessarily need to make you believe a particular viewpoint. It only needs to make you think, “This is a normal news article.”

When articles, journalist identities, and even media branding can all be mass-produced, traditional ways of evaluating information are put to the test.

AI doesn’t just lower the barrier to writing; it also lowers the cost of fabricating credible sources.

So what may truly become scarce in the future isn’t information, but credibility.

Before reading an article, in addition to asking what it says, we should also ask: Who produced it? Who published it? Who benefits from this narrative?