OpenAI Exposes Russian ChatGPT Network Running Fake Academic Institute

The Elaborate Deception

OpenAI dismantled a coordinated network of ChatGPT accounts it determined originated in Russia and systematically abused the chatbot to orchestrate an academic fraud scheme. According to OpenAI, the operators deployed Russian-language prompts to generate English-language social media posts and comments while employing virtual private networks to mask their geographic origin—a necessary workaround since ChatGPT is not available in Russia. The operation promoted the International Burke Institute, an organization falsely representing itself as an Israel-based research organization, using fabricated user profiles and impersonation tactics to distribute plagiarized content.

The scope of the plagiarism was staggering. OpenAI’s investigation revealed that 34 out of 36 articles examined on the institute’s website had been copied from legitimate sources, many misattributed to academics who did not author them. In one instance, an article addressing the China-Pakistan Economic Corridor came directly from Cambridge University Press but was falsely credited to a University of Nottingham professor specializing in South Asian affairs, creating false credentials to amplify credibility.

Growing Exploitation of Generative AI for Information Warfare

While individual social media posts from the campaign attracted minimal engagement, affiliated Telegram channels accumulated audiences reaching 20,000 followers. The sophistication of this operation distinguishes it from previous Russia-linked influence campaigns, according to OpenAI’s analysis. The pattern reflects an accelerating trend of state-sponsored actors weaponizing large language models for disinformation purposes.

This campaign fits within a documented history of similar operations. OpenAI identified a separate Russia-linked network called Doppelganger in 2024, which similarly leveraged ChatGPT to produce content praising Russia while simultaneously attacking Ukraine, NATO, and the United States across multiple platforms. Additionally, Forest Blizzard, a group connected to Russian military intelligence services, utilized ChatGPT for reconnaissance activities, script creation, and coordination of phishing operations.

Implications for Trust in Digital Spaces

Although OpenAI declined to formally attribute the International Burke Institute campaign to the Russian government, the technical sophistication and operational structure strongly suggest state-level coordination. The company noted that certain individuals based in Israel may have participated in representing the institute at conferences or submitting articles, though their precise relationship to the Russian operators remains unestablished.

The revelation underscores how generative AI technologies enable malicious actors to scale disinformation at velocity previously impossible through manual operations. For crypto markets—perpetually targeted by influence campaigns, impersonation schemes, and coordinated manipulation—these findings carry acute relevance. The ability to rapidly generate convincing false content and fabricate institutional authority directly threatens the due diligence processes investors and communities rely upon. Given crypto’s documented vulnerability to coordinated social engineering and information warfare, heightened skepticism toward unverified sources and institutional credentials has become operationally critical for market participants.

Source: OpenAI, via Decrypt. Not financial advice.