OpenAI Uncovers Russian Fake Israeli Research Institute Using ChatGPT for Influence Campaign

OpenAI revealed a covert pro-Russian influence operation that created a fake Israeli research institute to spread content supporting Russia and attacking Ukraine and the European Union. The operation used ChatGPT to generate articles, posts, and responses distributed across platforms like X, Facebook, LinkedIn, Telegram, and Substack. OpenAI blocked multiple ChatGPT accounts linked to the campaign after detecting AI-generated content.

The fake institute, alternately named International Anti-Corruption Institute (IAAI) and International Burke Institute (IBI), presented itself as a legitimate Israeli research body with a professional website, academic articles, expert lists including Noam Chomsky and Francis Fukuyama, and an international index favoring Russia. However, investigations found that 34 of 36 articles were plagiarized from other sources, and the institute was reportedly connected to Russian government interests. The campaign used VPNs to bypass ChatGPT access restrictions in Russia and crafted messages to conceal their Russian origin.

This operation demonstrated how AI tools can rapidly produce large volumes of seemingly credible content, combining it with fabricated institutional backing to lend false legitimacy to political messages. OpenAI’s investigation was led by Ben Nimo, who oversees efforts to detect AI misuse for disinformation. This is not the first time OpenAI has encountered pro-Russian actors exploiting ChatGPT; earlier in 2025, accounts linked to the pro-Russian Rybar media group were also blocked.

Other tech companies and researchers have noted similar trends, with Google reporting increased AI use by pro-Russian entities for influence campaigns, and Graphika identifying AI-generated content from Russian and Chinese sources. Social media platforms like YouTube, LinkedIn, and Snapchat have started implementing measures to label or reduce exposure to synthetic content, especially disinformation.

Israeli research group FakeReporter warned that AI enables political actors to fabricate authoritative appearances and manipulate public opinion at unprecedented scale and speed. With Israeli elections approaching, they stressed the urgent risk posed by AI-driven influence campaigns and the current regulatory and detection tools lagging behind the technology’s sophistication.