Russian Propaganda Ring Exposed: FBI Seizes Control of AI-Driven Bot Farm on X

Washington, United States – The United States, along with several allies, announced on Tuesday the successful seizure of a sophisticated Russian propaganda operation that utilized artificial intelligence to control nearly one thousand covert accounts on the social network X.

Using artificial intelligence to spread messages more widely and credibly, this takedown is noteworthy due to its connection to a Russian FSB intelligence officer and a former senior editor at state-controlled publication RT, revealed in court documents.

In a detailed joint advisory, agencies in the United States, Netherlands, and Canada outlined the various software programs used to manage the network, including Meliorator, which generated fake users in multiple countries. The FBI obtained a court order allowing the seizure of two web domains linked to the operation.

“FBI Director Christopher A. Wray described the operation as a groundbreaking effort to disrupt a Russian-sponsored social media bot farm enhanced by Generative AI,” as per a statement released.

The system evaded X’s verification techniques by automatically copying one-time passcodes sent to registered email addresses. The operation’s program code included references to Facebook and Instagram, indicating plans to expand to those platforms, according to officials.

The agencies recommended that social media companies enhance their methods for detecting covert automated behavior to prevent similar operations in the future.

X complied with a court order to provide information on the accounts to the FBI and subsequently deleted them. The Justice Department expressed gratitude to X for its cooperation in the investigation, signaling improved communication between the government and major social media companies.

Researcher John Scott-Railton of the Citizen Lab acknowledged the detailed information shared by the countries to aid other investigators and companies in identifying similar botnet operations.

Scott-Railton highlighted how AI’s large language models have facilitated Russian propagandists in expanding their reach and evading detection software. He emphasized the ongoing threat posed by such operations, noting that the current takedown represents only a fraction of the larger issue.