OpenAI says AI false-front campaigns reached real media outlets

OpenAI says it banned two influence operations that used ChatGPT to support false-front entities and place geopolitical messaging in front of real audiences. The company attributed one operation to Russia-origin actors and another to Iran-origin actors.

The Iranian operation used seven fake journalist personas to pitch long-form articles to small and medium online publications. OpenAI says that campaign reached Category 4 on the Breakout Scale because some material moved beyond social posts and appeared in external media outlets.

The Russia-origin operation was assessed as Category 5, the highest-impact influence operation OpenAI says it has disrupted so far. According to the company, operators used ChatGPT for internal reports, content work and management materials tied to a Latin America-facing research front.

The practical takeaway is blunt: generative AI did not invent the playbook, but it made old influence tactics faster, smoother and easier to localize. For publishers, platforms and readers, the weak point is no longer only synthetic posts. It is fake people and organizations using AI-polished material to pass through ordinary editorial channels.