THE HOOK
Russian influence operatives used ChatGPT to write progress reports for their bosses, and those reports are exactly how OpenAI caught them.
THE GIST
OpenAI says it banned Russian and Iranian accounts running influence operations across Latin America, the US and the UK. Iranian operators invented seven fake authors who placed almost 100 articles in real outlets, while a Russian crew impersonated a Peruvian education authority to stage school events.
THE TAKE
The scary bit isn't fake tweets, because nobody reads those. It's laundered bylines landing in real publications whose community sections skip editorial review. AI made the prose good enough to pass, and lazy gatekeeping did the rest, while the spies got caught doing paperwork. Bureaucracy wins again.
OPINION
NPR's write-up notes the Peru stunt ended with a Polish MEP reacting to a staged event, a real-world response to a fake story, which is the metric these shops actually sell. OpenAI's disclosures are useful, but they only catch operators careless enough to use a monitored US chatbot; anyone running open-weight models locally files no reports anywhere.
- Story type: Important story
- Impact: Medium
- Confidence: High (official announcement)
- Audience: Business
- What changed: OpenAI disclosed and banned Russian and Iranian accounts using ChatGPT for influence operations that placed content in real media outlets.
- What to watch: Whether outlets like Daily Kos and Middle East Monitor tighten community-section vetting after fake AI-polished bylines were exposed.
- Reading time: 2 minutes
