OpenAI Disrupts Russia-Origin AI Influence Campaign Using Fake Think Tank
OpenAI disclosed enforcement action against Russia-origin accounts tied to a newly reported covert influence campaign that weaponized ChatGPT alongside a synthetic Israel-based think tank identity. The IBI brand, registered in February 2025, served as a credibility shell while generative tooling churned social posts driving traffic to on-site articles and a sovereignty index framing Russia favorably against Western institutions. Investigators describe multi-channel amplification, concealed Russian account provenance, and reliance on scraped web material layered with a limited set of authentic Israeli-linked personas. The disruption highlights an evolving playbook: registered brand fronts plus AI-scaled distribution can lower barriers to laundering state-aligned narratives into Western political conversation with stronger plausible deniability than traditional sock networks. Readers should note that platform self-reporting carries verification limits, and independent attribution of state sponsorship remains contested in influence operations of this type.
OpenAI Disrupts Russia-Origin AI Influence Campaign Using Fake Think Tank
OpenAI banned Russia-origin accounts that used AI to promote a fake Israel-based think tank and a sovereignty index praising Russia and criticizing the West. The company published the disruption as part of a new covert influence campaign from Russia.
Key takeaway
State influence actors are pairing registered brand shells like IBI with ChatGPT-driven social amplification to inject pro-Russia narratives into Western politics at lower cost and higher deniability.
What happened
OpenAI reported banning Russia-origin accounts that used ChatGPT as part of a covert Russian influence operation promoting a fake Israel-based think tank brand called IBI, registered in February 2025, and a sovereignty index that praised Russia and criticized the West.
According to OpenAI, the operation's primary use of ChatGPT was generating social media posts to promote articles on the IBI website, amplified across multiple channels while concealing Russian user origins and drawing on copied web content plus a handful of real individuals in Israel.
Evidence
OpenAI banned Russia-origin accounts using AI to promote a fake Israel-based think tank and a sovereignty index praising Russia and criticizing the West.
OpenAI News · attributed
OpenAI banned Russia-origin accounts using AI to promote a fake Israel-based think tank and a "sovereignty" index praising Russia and criticizing the West.
The covert operation centered on a brand called IBI, registered in February 2025.
OpenAI News · attributed
OpenAI reports a covert Russian influence operation built around a brand called IBI, registered in February 2025
ChatGPT was primarily used to generate social media posts promoting articles on the IBI website.
OpenAI News · attributed
whose primary use of ChatGPT was generating social media posts to promote articles on the IBI website
The operation used multiple channels to amplify content and hid the Russian origins of its ChatGPT users.
OpenAI News · attributed
The operation used multiple channels to amplify content, hid the origins of its ChatGPT users in Russia
The operation drew on copied web content and a handful of real individuals in Israel.
OpenAI News · attributed
appears to have drawn on copied web content plus a handful of real individuals in Israel
Why it matters
Detection and moderation stacks must shift from evaluating isolated AI-generated posts toward provenance tracing, coordination patterns, and brand-front infrastructure that state-aligned operations increasingly reuse.
Limits and uncertainties
The available excerpt is a fragmented mix of attribution claims and commentary, making independent verification of core facts difficult.
Attribution of influence campaigns to specific state actors remains difficult even when platforms disclose disruption details.
The connection between specific narrative claims and the IBI operation is unclear in the fragmented published material.
Practical implications
Content-moderation and detection systems should treat AI-generated amplification behind registered brand fronts as a rising pattern requiring provenance and coordination signals rather than relying on single-post analysis.
Operators should monitor for defunct-looking brand fronts combined with AI-generated social content used to launder state-aligned messaging into Western political discourse.
What to watch
Whether similar registered brand-front operations emerge using generative AI to scale social amplification across multiple channels.
Follow-up OpenAI disclosures on coordinated inauthentic behavior and covert influence campaigns tied to Russia-origin accounts.
Platform detection efficacy against operations that conceal ChatGPT user origins while layering copied web content with authentic personas.
