This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
Russian propaganda channels utilized deepfake videos featuring Ukrainian Deputy Prime Minister Olha Stefanishyna to spread false claims about mandatory mobilization of women and travel restrictions. The Center for Countering Disinformation confirmed these videos were part of a coordinated disinformation operation designed to incite panic and destabilize the Ukrainian home front.
In late June 2025, Russian Telegram channels reportedly circulated deepfake videos claiming that Deputy Prime Minister Olha Stefanishyna backed mandatory mobilization of up to one million Ukrainian women starting September 1. Officials reportedly debunked the claim, confirming no such plans or laws exist. The disinformation operation reportedly aimed to incite panic and destabilize Ukraine's domestic situation.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
Using AI systems to conduct large-scale disinformation campaigns, malicious surveillance, or targeted and sophisticated automated censorship and propaganda, with the aim of manipulating political processes, public opinion, and behavior.
Also classified as
The campaign relied on generating and spreading false deepfake videos and fabricated claims to mislead the public.
The incident involves a coordinated, state-aligned disinformation campaign using synthetic media to influence public opinion and destabilize a population at scale.
Human
Due to a decision or action made by humans
Intentional
Due to an expected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed
The incident was driven by human actors (Russian propagandists) intentionally deploying deepfake technology to achieve a specific geopolitical goal.
Added categories peak at 3: Substantial (Epistemic Harm, Inferred)
Russian propaganda channels deployed deepfake videos of Ukrainian Deputy PM Olha Stefanishyna to falsely claim mandatory mobilization of young women. This coordinated state-aligned information operation aimed to destabilize the home front and erode trust in government institutions, representing a severe information warfare threat during an active conflict.
The AI system was used as a tool to generate synthetic video and audio of a public official. It performed a specific, deterministic function of deepfake generation based on direct inputs without exhibiting autonomous planning or agentic behavior.
Limited Risk: The report describes the use of AI-generated content (deepfakes), which falls under Limited Risk (Level 3) and requires specific transparency obligations to ensure users are informed of the AI's nature.
The reports explicitly state that Russian Telegram channels spread deepfake videos of the Deputy Prime Minister to disseminate false information.
The reports clearly outline the nature of the incident: a state-sponsored disinformation campaign using deepfakes of a specific political figure to sow panic. The role of AI (deepfake video generation) is explicit, and the political context is well-defined. There is little ambiguity regarding the intent or the primary domain classification.
None. The incident is a clear-cut case of geopolitical disinformation using synthetic media.
Technology
Deepfake Generation Tools
AI tools enabled creation of realistic fake videos of the Deputy Minister.
Algorithmic Amplification
Telegram algorithms facilitated rapid spread of the deepfake video.
Data Inputs
Source Video Exploitation
Authentic media of the official was used to train the deepfake model.
Human Factors
User Credulity
Panic led users to believe and share unverified synthetic videos.
Process and Methods
Inadequate Platform Moderation
Telegram lacked proactive detection and removal of malicious deepfakes.
Management
Adversary Information Warfare
Russian propagandists managed the deployment of AI to sow domestic panic.
Regulatory Environment
Lack of AI Governance
No strict regulations exist to prevent hostile state use of deepfakes.
AI system involved
An AI system was involved, but the reports do not identify it.
Identified by automated extraction; can differ from the alleged developers above.
