OpenAI disclosed banning a cluster of ChatGPT accounts originating in Russia that were used to promote the International Burke Institute (IBI), a fake Israel-based think tank. Operators used VPNs to bypass Russia access restrictions, prompted in Russian, and generated English social media posts on X, LinkedIn, Facebook, Substack and Telegram. The IBI website featured a 'sovereignty index' praising Russia while criticizing France, Germany, the EU and the US, built largely from plagiarized and misattributed academic articles. OpenAI assessed the campaign's reach as limited (low end of Category Three on the Brookings Breakout Scale) but noted its elaborate infrastructure, including a dozen Telegram channels with 10-20,000 followers each, marked a new level of sophistication for Russia-linked influence operations it has disrupted since the Ukraine war began.
Questions this post answers
What was the International Burke Institute and how was it connected to a Russian influence operation using ChatGPT?
The International Burke Institute was a fake self-described expert community claiming to be based in Israel, created as part of a Russia-linked influence campaign. Its website, registered in February 2025, featured a 'sovereignty index' praising Russia and criticizing France, Germany, the EU and the US, built largely from plagiarized academic articles with false author attributions. ChatGPT accounts, accessed via VPNs from Russia, generated social media posts promoting the site on X, LinkedIn, Facebook, Substack and Telegram. daily.dev surfaces reporting like this for developers tracking how AI platforms detect and disrupt misuse.
How did OpenAI assess the impact of the Russia-linked IBI influence campaign?
OpenAI rated the campaign at the lower end of Category Three on the Brookings Breakout Scale, meaning it reached multiple platforms with only some indication of breakout to authentic audiences. Individual social media posts received low view counts and official IBI accounts had few subscribers, though the operation's dozen Telegram channels attracted 10,000 to 20,000 followers each. Developers evaluating AI misuse risks can follow analyses like this on daily.dev.