1 hr ago
OpenAI Disrupts Russian and Iranian ChatGPT Propaganda Operations
OpenAI said it shut down two groups of accounts that used ChatGPT for propaganda work.
It linked one operation to Iran and another to Russia.
The Iran-linked operation used seven made-up journalist identities.
OpenAI found nearly 100 articles connected to those identities on about a dozen websites.
The operation also created social media comments about the US-Iran war.
The Russia-linked group used a supposed think tank called the Social Research Center.
People working for it in Latin America may not have known it was connected to a Russian group, according to OpenAI.
OpenAI said the Russian-linked activity aimed to harm Ukraine's reputation and affect local politics.
The articles do not include responses from Russia or Iran.
OpenAI said it banned ChatGPT accounts linked to two influence operations it associated with Russia and Iran.
The Iran-linked operation used seven purported Western journalist profiles to place articles in media outlets.
OpenAI identified almost 100 articles across roughly a dozen online outlets, dated from July 2025 to October 2026.
The Iranian operation also generated social media comments about the US-Iran war; accounts associated with one persona appeared suspended.
The Russia-linked operation used the purported Social Research Center and Latin American workers to target Ukraine's reputation and local politics.
- Who
- OpenAI reported disrupting operations it linked to Russia and Iran.
- What
- The operations used ChatGPT accounts to support false-front entities, fake journalist profiles, and covert propaganda activity.
- Where
- The operations targeted audiences internationally, including through outlets worldwide and activity in Latin America.
- When
- OpenAI made the announcement on Thursday; it identified Iranian-operation articles published from July 2025 through October 2026.
- Why
- OpenAI said the activity involved geopolitical messaging, including efforts to undermine Ukraine's reputation and influence local politics.
OpenAI's findings
Limitations and responses
Attribution and employee awareness
OpenAI's findings
OpenAI said the operations were linked to Russia and Iran, and that evidence indicated Latin American SRC employees did not know they were working for a Russian group.
Limitations and responses
The articles do not include responses from Russia, Iran, or the Social Research Center, nor independent confirmation of OpenAI's attribution.
Key facts
- Disrupting company
- OpenAI
- Operations
- Two, linked by OpenAI to Russia and Iran
- Iranian operation
- Used seven purported journalist profiles
- Articles identified
- Almost 100 published or syndicated across roughly a dozen online outlets
- Article dates identified
- July 2025 to October 2026
- Russian operation
- Used the purported Social Research Center, staffed by people in Latin America
Quotes
OpenAI
The AI company that investigated and disrupted the influence operations.
“We identified almost 100 articles published or syndicated under the operation's bylines across roughly a dozen online outlets around the world.”
NDTV
wionews.com
“The available evidence indicates that the SRC's employees in Latin America were not aware that they were working for a Russian group.”
NDTV
wionews.com
Sources
Related news
OpenAI Says Iran-Linked Operatives Used ChatGPT to Place Articles
Anthropic Reports AI-Assisted Spying by China and Iran
Anthropic Says Iran-Linked Users Targeted US Naval Forces
Anthropic Reports Claude Used for Weapons and Influence Operations
Researchers Warn of Fresh Russian Midterm Election Meddling
Anthropic Reports UAE-Linked AI Campaign Targeting Sudan War Narratives
Former German Chancellor Schroeder Celebrates Putin’s Birthday at Home
Putin Pledges Support to Pezeshkian for Ending War
Zelenskiy Accuses US of Inaction After Deadly Russian Strikes
Denmark Says Russia Tried to Sabotage Ukraine-Supplying Defence Firms
