OpenAI Bans Accounts Tied to Russian and Iranian Influence Operations

OpenAI announced in a blog post that it banned Russian and Iranian accounts utilizing its artificial intelligence tools for influence operations across multiple countries. According to OpenAI, internal reports generated by the operatives using ChatGPT indicated that some false narratives were picked up by news outlets and even prompted a response from a member of the European Parliament.

Readers can verify official details by reviewing OpenAI's published company blog posts and statements from independent researchers or affected media organizations.

Key Findings from OpenAI's Investigation

• Russian Operations in Latin America: Russian accounts targeted Latin American nations to undermine support for Ukraine and influence local politics, primarily using ChatGPT to draft internal status reports for superiors.
• The Peru and Poland Incident: Operatives impersonated a Peruvian educational authority, tricking schools into holding events featuring Stepan Bandera, which led to a Polish news report and subsequent comments by a far-right Polish member of the European Parliament.
• Iranian Content Generation: Iranian accounts used AI to assist in creating and refining content in Persian and English, focusing primarily on the U.S.-Iran conflict and American politics.
• Media Placement: Iranian accounts utilized fake author personas to pitch and publish nearly 100 articles in various international outlets, including Daily Kos and Middle East Monitor.

Practical Implications and Verification

Independent experts note that AI tools can help threat actors generate authentic-sounding language, presenting ongoing challenges for media organizations regarding submission reviews and editorial safeguards. Independent researchers and media platforms continue to review their processes to better detect and prevent the misuse of publishing tools.