YouTube as a propaganda platform: analysis of Belarusian state media coverage of the Zapad 2025 military exercises

Key findings of the study / TL;DR
This publication is a continuation of our previous analysis.A comparative study of coverage of the Zapad-2021 and Zapad-2025 exercises in Belarusian Telegram channelsThis analysis is based on our regular monitoring of key state YouTube channels with over 1 million subscribers. Three of these five are YouTube versions of television channels. The other two are the main news agency and a major publishing house.
Our dataset for this year contains over 13,500 videos, 271 of which (2%) mentioned the joint Belarusian-Russian military exercises “Zapad-2025.” All transcribed versions of these videos were processed using LLM models for risk level, disinformation narratives, and coordinated campaigns.
50 claims require additional verification (classical fact-checking). Three coordinated campaigns were detected. The predominant narrative technique is military_rhetoric. About a third of the materials are classified as high-risk.

Methodology and limitations
Basic Analysis: Statistical analysis of video metadata and temporal patterns. AI Analysis: Using gpt-4o-mini to detect disinformation techniques. Campaigns: Finding coordinated activity based on temporal and content patterns.
Strategic Sampling: Uniform time distribution for AI analysis.

Limitations:
Analysis is based on automatically extracted fragments of mentions. AI assessments require additional expert verification. Coordination patterns may have innocent explanations. Sample size for AI analysis is limited by API quotas. Focus on quantitative metrics; qualitative analysis requires expertise.

A detailed analysis of disinformation techniques

Risk distribution and intensity of disinformationAn analysis of risk levels reveals a disturbing picture: 38.1% of materials are classified as high-risk (7-10 points), representing the largest share. Medium-risk content (4-6 points) accounted for 36.1%, while low-risk content accounted for only 25.7%. This pyramidal structure, inverted toward high risk, indicates a deliberate informational influence strategy rather than random journalistic errors.
A graph of the intensity of disinformation techniques by channel reveals the leaders in propaganda activity. The STV (Capital Television) channel demonstrates the highest scores for all key techniques: military_rhetoric reaches 4.6, false_facts 3.3, and emotional_manipulation 1.8. The BelTA news agency shows moderate scores (around 1.1-1.3 for the main techniques), which can be explained by the news agency’s format, which requires at least superficial adherence to factual standards.

Dominant narratives and target audiences
The distribution of mentions of countries and organizations reveals the geopolitical focus of the propaganda. Russia is mentioned in over 80 materials, as expected for joint exercises. However, the significant attention to Ukraine (approximately 50 mentions) and NATO (approximately 30 mentions) indicates an attempt to create a narrative of an external threat. Poland and Western countries also appear in a significant number of materials, creating an image of a hostile environment.
The emotional tone of the content demonstrates extreme polarization. About 200 materials have a negative tone, while only about 70 are positive, and fewer than 50 are neutral. This 3:1 ratio in favor of negativity indicates a deliberate attempt to create an atmosphere of fear and hostility, a classic mobilization propaganda technique.

Coordinated activity and systemic nature
The discovery of three coordinated campaigns confirms the hypothesis of centralized control over the information flow. The predominance of the military_rhetoric technique, the classification of a third of the materials as high-risk, and the presence of temporal patterns in publications indicate that coverage of the Zapad-2025 exercises is not journalistic work, but rather part of an information operation aimed at mobilizing the population, legitimizing militarization, and shaping the image of a hostile environment.

The table of the top 15 most dangerous content reveals characteristic patterns. STV dominates the top positions with materials receiving the highest risk ratings (9.0) and conspiracy theories (9.0). Headlines like “Weekdays 11.04.2025. FULL VERSION” and “Drang nach Osten: Europe is building a new Reich” demonstrate the use of historical analogies and apocalyptic narratives. Articles about “The Baltics sold out” and “Land and people become hostages” point to a strategy of dehumanizing neighboring states.
NEWS.BY (Bel news) demonstrates a slightly different approach, with risk ratings of 8.0 and higher scores for false facts (8.0), indicating the specific nature of its work with factoids and pseudo-documentary materials.

Bank of statements for fact-checking
Fifty allegations were identified that require priority review. The allegations are sorted by risk level and potential impact.

Fifty high-priority claims identified reveal the structure of the disinformation campaign. The most common themes include threats from NATO, Poland, and Ukraine, conspiracy theories about plans for attack, demonization of Western countries, and glorification of Belarusian-Russian military cooperation.
Claims about “Trump yielding to Russia,” “war with China,” and “Europe building a new Reich” are aimed at creating the illusion of a global conflict in which Belarus and Russia act as defenders against aggression. Particularly revealing are the geographic specifics: “30,000 NATO troops near the borders of Belarus,” “strike range up to 300 km,” “nuclear warheads for the Polonaise.” These pseudo-accurate data are intended to create an illusion of credibility in the absence of verifiable sources.

Priority actions:
Urgent Verification: Immediately verify 10 materials with the highest risk. Campaign Monitoring: Establish enhanced surveillance of channels with coordinated activity.
Fact Checking: Verify 50 high-priority claims.

Long-Term Recommendations: Expand automated monitoring of identified channels. Create an early warning system for new campaigns. Develop cooperation with international organizations working with FIMI/DIMI. Regularly update disinformation pattern databases. Conduct training for journalists on identified techniques.

Additional Material:
Structured dataset