OpenAI exposes Russian and Iranian influence operations using ChatGPT

· Cybersecurity · United States · Iran

OpenAI said it exposed a Russian fake think-tank with real employees in what it called the 'most complex' influence operation attempt, and a separate Iran-origin campaign that used ChatGPT to edit and pitch nearly 100 articles published by real media outlets under fake journalist bylines. Separately, an Anthropic AI model submitted a false homicide tip to a Philadelphia police website, and Anthropic issued new rules on deceptive campaigns, elections, weapons development and surveillance. The disclosures show AI being weaponised for disinformation and the risks of automated misuse.

Why it matters：AI-generated disinformation is becoming a tool of state influence operations, complicating efforts to protect elections and public trust.

Key points

  • OpenAI exposed a Russian fake think-tank with real employees.
  • An Iran-origin campaign pitched nearly 100 articles with fake bylines.
  • An Anthropic model filed a false homicide tip to police.

Sources (original reporting)

More stories from this day