OpenAI Bans Russian Accounts Behind Fake Think Tank

OpenAI disrupted a covert Russian influence operation that used ChatGPT to construct the appearance of an independent, credible research institution, ultimately banning a cluster of accounts the company assessed as very likely originating in Russia, according to OpenAI's own report on the disruption, published Tuesday.

The campaign centered on the International Burke Institute, a self-described "expert community" claiming to be based in Israel, whose website was registered in February 2025 and featured prominent academics and published research designed to give the operation genuine institutional credibility. The illusion was largely manufactured. Of 36 articles the institute published on its site between September 2025 and May 2026, 34 had been copied from elsewhere online, according to IBTimes' reporting on OpenAI's findings.

The operators accessed ChatGPT through VPNs, since OpenAI doesn't make its models available directly in Russia, and typically prompted the system in Russian while generating mostly English-language content, according to CNBC's reporting on the disclosure. Crucially, operators explicitly instructed ChatGPT to remove any linguistic clues that could reveal their Russian origin before publishing the generated comments across Substack, Telegram, X, Facebook, and LinkedIn.

How the Campaign Was Structured

A central feature of the operation was a "sovereignty index," a purported analytical ranking that portrayed Russia favorably while denigrating Western countries by comparison. According to reporting from The Decoder, the index rated Russia's military half a point higher than the United States military, a fabricated metric designed to lend quantitative-sounding legitimacy to a pro-Kremlin narrative. The institute published reports specifically targeting France, Germany, the United States, and the European Union, while ChatGPT-generated social media posts were used to direct readers toward that material.

The network's reach went further than the institute itself. One operator used ChatGPT to generate German-language posts for a Telegram channel that criticized Ukraine, the EU, and the German government while advocating closer ties with Russia. Separately, ChatGPT was used to design logos for roughly a dozen Telegram channels targeting audiences in the United States, France, Germany, Poland, and Türkiye, with individual channels amassing between 10,000 and 20,000 subscribers each, according to OpenAI's report.

OpenAI was candid that this specific campaign's actual audience remained limited. "Although the campaign appears to have reached relatively small audiences, its elaborate construction distinguishes it from other Russia-linked influence operations we have disrupted," the company said in its report. Researchers emphasized the real risk lies in the infrastructure itself, which could be scaled up significantly over time, a concern that connects directly to a broader pattern we've tracked closely, including OpenAI's earlier disruption of a campaign dubbed "Data Center Bandwagon" that spread false claims linking AI data centers to rising household electricity costs, and separate Russian influence operations previously reported targeting African elections with budgets reaching $600,000 annually.

This disclosure is worth understanding for any business operating in media, communications, public relations, or content moderation, since it demonstrates how convincingly AI tools can now be used to manufacture the appearance of independent institutional credibility at relatively low cost. The specific technique here, building a seemingly legitimate think tank complete with plagiarized academic content and a fabricated analytical index, represents a genuinely more sophisticated approach than earlier, cruder disinformation efforts.

For businesses evaluating information sources, partnerships, or citing third-party research in their own content, this case is a reminder to verify the underlying credibility of unfamiliar "expert" institutions and indices before treating them as legitimate sources, particularly ones that emerged recently and lack a long-established public track record.

What was the International Burke Institute?
The International Burke Institute was a fake think tank claiming to be based in Israel, created as part of a Russian-linked influence operation that used ChatGPT to generate supporting content and used plagiarized academic articles to build false credibility.

How did OpenAI discover this influence campaign?
OpenAI discovered the operation while investigating suspicious AI-generated social media posts, which led researchers to uncover the broader network built around the fake institute, a fabricated "sovereignty index," and coordinated content across multiple platforms.

Did this Russian influence campaign actually reach a large audience?
No. OpenAI said the campaign's reach remained relatively small, though the company emphasized the sophisticated infrastructure built around it could be scaled up significantly in future operations.

OpenAI banned a cluster of ChatGPT accounts likely originating in Russia that built a fake Israeli think tank called the International Burke Institute, using AI to generate plagiarized-appearing research and a fabricated "sovereignty index" ranking Russia favorably against Western countries. Operators accessed ChatGPT through VPNs and specifically instructed it to hide linguistic clues of their Russian origin while distributing content across Substack, Telegram, X, Facebook, and LinkedIn. OpenAI said the campaign's actual audience reach remained small, but warned the sophisticated infrastructure built around it represents a genuine template that could be scaled up in future influence operations.