Purported Deepfake Videos Reportedly Impersonated Yanis Varoufakis on YouTube and Social Media

Purported deepfake videos using the face and voice of Greek economist and politician Yanis Varoufakis were reportedly circulated on YouTube and other social platforms in late 2025 and early 2026. The videos depicted a synthetic version of Varoufakis delivering fabricated political statements, including about international events. Platforms reportedly removed some videos, but many reappeared under new accounts.

Risk classification

• Primary risk domain: 4 Malicious actors
• Primary risk subdomain: 4.3 Fraud, scams, and targeted manipulation

The incident involves the unauthorized use of a public figure's likeness to create deepfakes for targeted manipulation, reputation damage, and political impersonation.

Additional risk subdomains

• 3.1 False or misleading information: The deepfakes generate false political statements, such as fictitious claims about a coup in Venezuela, misleading the public.
• 4.1 Disinformation, surveillance, and influence at scale: The deepfakes have proliferated at scale across major platforms to muddy public political discourse and drown genuine dissent.

Causal factors

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

The incident was caused by malicious human actors intentionally creating and distributing deepfake videos of the victim post-deployment.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The report describes the creation and proliferation of 'AI-generated content such as deepfakes', which falls under the transparency obligations of Limited Risk systems.

AI system and alleged parties

• AI system: AI chatbot, unspecified deepfake generator
• AI purpose: Deepfake Video Generation; Voice Generation
• Behaviour type: Tool
• Alleged developer: Unknown voice cloning technology developers, Unknown deepfake technology developers
• Alleged deployer: Unknown deepfake content creators impersonating Yanis Varoufakis, Unknown deepfake content creators
• Alleged harmed parties: YouTube users, Yanis Varoufakis, social media users, Epistemic integrity

Harm severity

Severity ratings are based on this scale.

Highest direct severity in any category: Substantial. 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 Substantial, indirect Negligible
• Differential treatment: direct Negligible, indirect Negligible
• Civil rights: direct Negligible, indirect Negligible
• Democracy: direct Minor, indirect Negligible
• Privacy: direct Negligible, indirect Negligible
• Psychological: direct Minor, indirect Negligible
• Epistemic: direct Minor, indirect Negligible
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Malicious content

Reported: Yes, the report describes hundreds of malicious deepfake videos created to impersonate the victim.

Directly caused: Hundreds of deepfake videos synthesizing the victim's voice and face were created and spread across YouTube and social media.

Inferred additional harm: It is likely that similar deepfake campaigns are targeting other public figures, though not detailed in this report.

Democracy

Reported: Yes, the report describes how deepfakes are used to muddy discourse and drown genuine dissent.

Directly caused: The deepfakes depicted the victim making fictitious statements about political events like the coup in Venezuela, muddying public political discourse.

Inferred additional harm: The proliferation of unverified synthetic media of political figures threatens to erode public trust in democratic discourse and authentic communication.

Psychological

Reported: Yes, the report describes psychological distress, rage, and a feeling of loss of self-ownership.

Directly caused: The victim experienced initial rage, anxiety, and a sense of being a spectator to his own digital puppetry, feeling that his audiovisual identity was stolen.

Epistemic

Reported: Yes, the report describes the generation of false political statements and the erosion of the ability to verify who is speaking.

Directly caused: The deepfakes generated false and misleading political claims, causing supporters to question what was real and opponents to weaponize the fake statements.

Inferred additional harm: The widespread availability of highly persuasive deepfakes contributes to a broader loss of shared consensus reality and increased skepticism toward all digital media.

People affected

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

Potential causes

Management

• Prioritizing Engagement Over Safety: Platforms focus on cloud rents and traffic rather than proactive verification.
• Absence of Proactive Verification: Platforms lack automated systems to detect and block deepfakes before upload.

Technology

• AI Voice and Face Synthesis: Creates highly persuasive synthetic videos of individuals without their consent.
• Algorithmic Recommendation Feeds: Amplifies deepfakes by prioritizing engagement over content authenticity.

Data Inputs

• Publicly Available Media Data: Original videos are harvested to train and clone the victim's voice and face.

Human Factors

• Audience Confusion: Viewers struggle to distinguish between real and deepfaked political content.
• Weaponization by Opponents: Opponents circulate synthetic videos to damage the victim's reputation.

Process and Methods

• Slow Platform Takedown Processes: Reporting processes take a week or more to remove infringing deepfake videos.
• Ineffective Reappearance Barriers: Banned channels instantly recreate profiles to re-upload deepfake content.

Regulatory Environment

• Lack of Digital Identity Protections: No legal framework prevents the unauthorized AI replication of human identity.

Information quality

• Classification confidence: High
• Reason for confidence: The report is a first-hand account by the victim, providing clear details about the nature of the deepfakes, the platforms used, the impact on his reputation, and his attempts to have them removed. While the specific AI models are not named, the description of the synthetic media and its effects is highly detailed and unambiguous.
• Ambiguities identified: The specific AI tools or creators of the deepfakes are not identified.

A persistent deepfake campaign targeted Greek politician Yanis Varoufakis with fabricated political statements, causing public confusion and reputational harm. While demonstrating the evolving challenge of synthetic media and platform moderation, the incident has minor national security implications and did not disrupt critical sovereign functions.

• Overall national security impact: Minor
• Scope: Single nation
• Primary target: Greece
• Alleged perpetrator: Unknown

Threat characteristics

• Imminence: Long-term. Represents an ongoing strategic concern regarding the proliferation of synthetic media rather than an immediate, time-sensitive national security crisis.
• Autonomy: Human-controlled. The AI systems were used as generative tools by human actors who directed the content, creation, and dissemination of the deepfakes.
• Novelty: Evolved capability. While political deepfakes are an established threat, the high volume, rapid replication, and persistence against platform moderation represent an evolved challenge.

Impact by dimension

• Physical security: Negligible. No physical threats, kinetic actions, or critical infrastructure impacts were reported in connection with this deepfake campaign.
• Information security: Minor. The campaign involved hundreds of deepfakes spreading fabricated political statements, but it lacked clear attribution to a foreign state adversary or impact on national intelligence.
• Sovereignty: Minor. While the deepfakes targeted a prominent political figure and sowed public confusion, they did not disrupt core government operations, borders, or electoral processes.
• Economic security: Negligible. The incident did not target financial systems, strategic technologies, or critical supply chains.
• Societal stability: Minor. The deepfakes caused public confusion and personal psychological distress to the victim, but did not lead to widespread civil unrest or systematic human rights violations.