Updated
Updated · OpenAI · Aug 25
OpenAI Bans Russian ChatGPT Accounts in New Influence Campaign, Reaching 10,000-20,000 Telegram Followers
Updated
Updated · OpenAI · Aug 25

OpenAI Bans Russian ChatGPT Accounts in New Influence Campaign, Reaching 10,000-20,000 Telegram Followers

3 articles · Updated · OpenAI · Aug 25

Summary

  • OpenAI said it removed a cluster of Russia-linked ChatGPT accounts that generated English and German posts promoting the International Burke Institute, a purported Israel-based think tank, across X, Facebook, LinkedIn, Substack and Telegram.
  • The accounts used VPNs to bypass OpenAI’s Russia access ban and prompted in Russian while asking ChatGPT to hide linguistic signs of Russian origin, according to the company’s investigation.
  • At the center was the IBI website, registered in February 2025; in a sample of 36 articles published between September 2025 and May 2026, OpenAI said 34 were copied or misattributed academic works.
  • The operation also pushed a “sovereignty index” that praised Russia and attacked Western countries, while one operator generated content for the German-language Telegram channel Lahme Ente and another designed logos for about a dozen channels.
  • OpenAI said the campaign’s direct reach was limited—most posts drew few views, though Telegram channels had roughly 10,000-20,000 followers each—but called the operation notable for its elaborate infrastructure and use of AI to manufacture credibility.

Insights

How did Russian operatives trick real individuals into supporting a fake AI-driven think tank without exposing their true identities?
What hidden flaws allowed OpenAI to uncover this sophisticated, VPN-cloaked disinformation network before it could cause widespread harm?
If AI can manufacture entire academic ecosystems, how can anyone verify the true origin of online expertise today?