Reportedly Viral USAID Disinformation Video Linked to Russian-Aligned Campaign Known as Matryoshka

A video that reportedly went viral on X in early February 2025 claimed USAID paid celebrities to visit Ukraine. The clip mimicked E! News branding and included a narrator with a British accent. Individuals and organizations named denied any involvement. Researchers attributed the video to Matryoshka (AKA Operation Overload and Storm-1679), an influence campaign suspected of using AI-generated narration and editing, though tool use has not been independently confirmed.

Risk classification

• Primary risk domain: 4 Malicious actors
• Primary risk subdomain: 4.1 Disinformation, surveillance, and influence at scale

The incident involved a coordinated, state-aligned influence campaign using AI-generated content to manipulate public opinion and political processes at scale.

Additional risk subdomains

• 3.1 False or misleading information: The AI system was used to generate a highly viral false video that misled prominent figures and millions of social media users.

Causal factors

• Entity: Human
• Intent: Intentional
• Timing: Post-deployment

The incident was driven by human actors who intentionally used AI tools post-deployment to generate and spread false narratives to undermine USAID.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The system generated deepfakes (AI-generated content mimicking real news and voices), which falls under transparency obligations for AI-generated content.

AI system and alleged parties

• AI system: unspecified
• AI purpose: Voice Generation; Deepfake Video Generation
• Behaviour type: Tool
• Alleged developer: Unknown voice cloning technology developers, Unknown deepfake technology developers
• Alleged deployer: Storm-1679, Russian-linked disinformation network, Operation Overload, Matryoshka
• Alleged harmed parties: USAID, Truth, The New York Times, Sean Penn, Politico, Orlando Bloom, New York University, E! News, Ben Stiller, BBC, Angelina Jolie

Harm severity

Highest direct severity in any category: Severe. Severity is scored from Negligible to Catastrophic in each harm category, for harm the reports describe as caused directly or indirectly by the AI system.

• Physical: direct Negligible, indirect Negligible
• Infrastructure: direct Negligible, indirect Negligible
• Property: direct Negligible, indirect Negligible
• Financial: direct Negligible, indirect Negligible
• Environmental: direct Negligible, indirect Negligible
• Malicious content: direct Minor, indirect Severe
• Differential treatment: direct Negligible, indirect Negligible
• Civil rights: direct Negligible, indirect Negligible
• Democracy: direct Minor, indirect Minor
• Privacy: direct Negligible, indirect Negligible
• Psychological: direct Negligible, indirect Negligible
• Epistemic: direct Minor, indirect Severe
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Malicious content

Reported: The report explicitly describes the creation and spread of a fake news video containing malicious disinformation.

Directly caused: The AI-generated voice-over was used to create a fake E! News video falsely accusing USAID of paying millions to celebrities.

Indirectly caused: The video was shared widely on X, Truth Social, and TikTok, reaching millions of users and fueling online anger and conspiracy theories.

Inferred additional harm: It is likely that this campaign is part of a larger network of hundreds of AI-facilitated propaganda posts targeting Western audiences.

Democracy

Reported: The report explicitly describes how the disinformation campaign aimed to disrupt domestic politics and weaken support for Ukraine.

Directly caused: The fake video was used by political figures to justify dismantling a government agency (USAID) and fuel conspiracy theories.

Indirectly caused: The campaign aligned with Kremlin goals to influence US foreign policy and public opinion regarding aid to Ukraine.

Inferred additional harm: Such campaigns likely contribute to a broader erosion of trust in democratic institutions and public administration.

Epistemic

Reported: The report explicitly describes the spread of AI-driven misinformation and the erosion of shared truth.

Directly caused: The fake video fabricated claims about USAID funding, misleading prominent figures like Elon Musk and Donald Trump Jr.

Indirectly caused: The viral spread of the video and subsequent false claims about media funding polluted the information ecosystem, making it harder for factual corrections to reach the public.

Inferred additional harm: Widespread use of cheap AI tools to generate convincing fakes likely leads to a general decline in public trust in legitimate news organizations.

People affected

• Occurrences reported: 1
• People reportedly harmed: 4
• People reportedly exposed: 4000000

Potential causes

Management

• Platform Research Tool Shutdown: Platforms disabled tools, hindering researchers from tracking disinformation.
• Prioritizing Engagement Over Safety: Platform management allowed debunked fake videos to remain online.

Technology

• AI-Generated Voice-Overs: AI voice-overs lowered production costs of credible-sounding fake news.
• Impersonation of Trusted Brands: AI tools allowed pairing real logos with synthetic audio to deceive viewers.

Data Inputs

• Lack of Content Provenance Data: Platforms lacked data to verify the origin of the E!News video clip.
• Pre-seeded Influencer Networks: Disinformation was fed directly to receptive pro-Russian social accounts.

Human Factors

• High-Profile Amplification: Elon Musk and others reposted the fake video, reaching millions.
• Confirmation Bias of Users: Audiences hostile to USAID accepted the fake video without verifying.

Process and Methods

• Inadequate Algorithmic Filtering: X failed to automatically flag or remove the synthetic media.
• Delayed Fact-Checking Mitigation: Community Notes added context only after the video gained millions of views.

Regulatory Environment

• Defunding of Tracking Agencies: Disbanding the Global Engagement Center weakened foreign threat monitoring.
• Pressure Against Content Moderation: Political pushback led platforms to scale back moderation of fake content.

Information quality

• Classification confidence: High
• Reason for confidence: The reports from reputable sources provide clear, consistent details about the specific fake video, its origins, its reach, and the AI techniques used. The role of the AI as a tool in a broader disinformation campaign is well-documented.
• Ambiguities identified: The specific AI models used to generate the voice-overs are not named.

A Russian-aligned influence campaign ('Operation Overload') used AI-generated voice-overs to create a fake news clip targeting USAID and US support for Ukraine. The incident represents a substantial information warfare and sovereignty threat, demonstrating how cheap AI tools can be leveraged by foreign actors to influence domestic political discourse and undermine public trust in government agencies.

• Overall national security impact: Substantial
• Response level: Substantial
• Scope: Multiple nations
• Primary target: United States
• Other affected: Ukraine
• Alleged perpetrator: Russian-aligned actors

Threat characteristics

• Imminence: Long-term. Represents an ongoing strategic concern regarding foreign state-aligned disinformation campaigns rather than an immediate physical crisis.
• Autonomy: Human-controlled. AI systems were used directly by human operators as a tool to generate synthetic voice-overs for edited video clips.
• Novelty: Evolved capability. While deepfakes are an established threat, the coordination with domestic political narratives and rapid high-profile amplification represents an evolved capability.

Impact by dimension

• Physical security: Negligible. No physical systems, critical infrastructure, or human safety threats were involved in this digital disinformation incident.
• Information security: Substantial. Coordinated Russian-aligned influence campaign used AI voice-overs to fabricate a news report, targeting US foreign policy and USAID, achieving millions of views and amplification by prominent figures.
• Sovereignty: Substantial. Foreign state-aligned actors used AI-generated content to influence US domestic political debates regarding the funding and survival of a federal government agency (USAID).
• Economic security: Negligible. No significant threats to financial systems, strategic technologies, or economic infrastructure were reported in this incident.
• Societal stability: Minor. The campaign aimed to manipulate public opinion and erode trust in public institutions, but did not result in civil unrest or physical human rights violations.