Threat actors affiliated with RT (formerly Russia Today), a Russian government-backed media organization, have used artificial intelligence (AI) features of the Meliorator software to create fake online personas used to disseminate disinformation to and about the US, Germany, Israel, the Netherlands, Poland, Spain, and Ukraine, reads a joint advisory from the government agencies.
Meliorator was designed to create seemingly authentic social media personas en masse, post content like authentic users, mirror disinformation posted by other fake personas, perpetuate false narratives, and formulate messages.
The tool consists of an administrator panel named Brigadir, which serves as the primary end-user interface, and a seeding tool named Taras, a back-end used to control the fictitious personas.
Threat actors could use Meliorator to create bots based on specific parameters or archetypes and had access to automated scenarios or actions that could be completed on behalf of groups of bots.
According to the authoring agencies, RT had access to the AI-enabled bot farm generation and management software since 2022 and used it to disseminate disinformation in support of Russia’s interests.
As of June 2024, Meliorator was only compatible with X, where Russian threat actors used it to create 968 accounts to use as part of their influence operations.
On Tuesday, the US announced it seized two domain names that were used by private email servers the Russian threat actors relied on to register the fake social media accounts for the bot farm.
The bot farm, the affidavits filed in support of the seizure warrants show, was controlled by a Russian FSB officer, a Meliorator developer, and members of a private intelligence organization, “with the approval and financial support of the Presidential Administration of Russia (aka the Kremlin)”.
The identified accounts have been suspended by X, but the government agencies call for other social media platforms to take note of the influence operation and “assist with identifying fictitious personas to reduce Russian malign foreign influence activity”.
Related: Google Disrupts More China-Linked Dragonbridge Influence Operations
Related: Why We Need to Get a Handle on AI
Related: Senators Urge $32 Billion in Emergency Spending on AI After Yearlong Review
