Dimal Basha's Likeness Was Reportedly Used in Purportedly Fake AI-Generated Video After Kosovo Parliament Speaker Election

Minutes after Kosovo MPs elected Vetëvendosje's Dimal Basha as Parliament Speaker, a purportedly AI-generated Facebook video used his likeness to claim he was the father of Kosovo's Ashkali community, a Roma-related minority group that has faced long-running discrimination and political marginalization. The video reportedly received more than 80,000 views in under eight hours.

Risk classification

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

The incident involves coordinated and at-scale dissemination of AI-generated deepfakes targeting political figures to influence public opinion and democratic processes in Kosovo.

Additional risk subdomains

• 3.2 Pollution of information ecosystem and loss of consensus reality: The rapid spread of unverified AI content on Facebook and TikTok is polluting the local media ecosystem and eroding public trust.
• 1.2 Exposure to toxic content: The deepfakes include degrading portrayals of political leaders, contributing to political polarization and hate speech.

Causal factors

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

The incident was caused by human actors intentionally using deployed AI generation tools to create and spread deepfakes to manipulate public opinion.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The report describes the generation of deepfakes, which under the EU AI Act are subject to transparency obligations to ensure users are informed that they are interacting with AI-generated content.

AI system and alleged parties

• AI system: unspecified
• AI purpose: Deepfake Video Generation; Voice Generation
• Behaviour type: Tool
• Alleged developer: Deepfake technology developers
• Alleged deployer: Disinformation actors in Kosovo, Disinformation actors, Deepfake creators
• Alleged harmed parties: National security and intelligence stakeholders, Minorities, General public of Kosovo, General public, Epistemic integrity, Electoral integrity, Dimal Basha, Democratic integrity, Ashkali people, Ashkali community in Kosovo

Harm severity

Severity ratings are based on this scale.

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

Malicious content

Reported: Yes, the report explicitly describes the creation and spread of degrading and falsified AI-generated videos.

Directly caused: At least 25 AI-generated videos on Facebook and over 100 on TikTok were identified, including degrading portrayals of political leaders.

Inferred additional harm: It is likely that thousands of additional malicious deepfakes and manipulated audio clips continue to circulate undetected across social media platforms in Kosovo.

Democracy

Reported: Yes, the report explicitly describes how AI-driven disinformation threatens democratic processes and election integrity.

Indirectly caused: The proliferation of deepfakes threatens to distort the electoral process, meaning formed governments may not reflect the real will of the electorate, and deepens political polarization.

Inferred additional harm: The ongoing manipulation of political discourse is likely to permanently erode public trust in democratic institutions and the legitimacy of future election outcomes in Kosovo.

Psychological

Reported: Yes, the report describes degrading portrayals of political leaders.

Directly caused: Political leaders like Dimal Basha, Albin Kurti, and Ramush Haradinaj were subjected to degrading portrayals and fabricated statements, causing reputational and potential emotional distress.

Inferred additional harm: It is likely that targeted political figures and their families experienced psychological distress due to targeted public humiliation, though specific details are not quantified.

Epistemic

Reported: Yes, the report explicitly describes the rapid spread of AI-generated misinformation threatening the credibility of information.

Directly caused: AI-generated videos, such as one falsely claiming a politician is the father of the Ashkali community, amassed over 80,000 views, spreading false beliefs.

Indirectly caused: The volume of deepfakes blurs the boundary between truth and falsehood, creating a 'hyper-reality' that undermines the public sphere.

Inferred additional harm: The widespread exposure to deepfakes is likely causing long-term erosion of shared reality and media trust among the Kosovo public, particularly older generations.

People affected

• Occurrences reported: 1
• People reportedly harmed: 4
• People reportedly exposed: 1000000

Potential causes

Management

• Platform Governance Failure: Social media companies fail to manage and remove harmful deepfakes.

Technology

• Rapid AI Generation Tech: Rapid development of AI fuels easy creation of manipulated video and audio.
• Highly Realistic Deepfakes: AI-generated content looks so real that audiences fall prey to propaganda.
• Platform Removal Failure: Social platforms fail to provide methods to remove AI-generated content.

Data Inputs

• Fabricated Audio and Video: Manipulated biometric data is used to synthesize fake political statements.
• Unlabeled AI Content: AI materials are shared by users without any clarification of being fake.
• Foreign Disinformation Feeds: Kremlin-linked networks feed disinformation to multiply hate speech.

Human Factors

• Low Media and Tech Literacy: Low technological literacy leads to high vulnerability to AI deepfakes.
• Vulnerability of Older Adults: Older generations take deepfakes seriously, reinforcing political beliefs.
• Unchecked Sharing by Users: Regular users share fake content thousands of times without verification.

Process and Methods

• Delayed Fact-Checking: By the time fact-checkers step in, the viral damage is already done.
• Lack of School Curriculum: Public school curriculum lacks media and artificial intelligence literacy.
• No Dedicated Verification: Influential media outlets lack dedicated teams to verify AI content.

Regulatory Environment

• No State Institutional Response: State institutions fail to provide a serious response to AI manipulation.

Information quality

• Classification confidence: High
• Reason for confidence: The reports provide clear, consistent accounts of the deepfake phenomenon in Kosovo, backed by specific statistics from certified fact-checking organizations (Krypometer, BIRN Kosovo) and quotes from local experts and political representatives. The role of AI as a tool for generating these deepfakes is explicitly stated.
• Ambiguities identified: The specific AI models or platforms used to generate the deepfakes are not identified.

The coordinated spread of AI-generated deepfakes targeting Kosovo's political leadership poses a substantial threat to information security, democratic processes, and societal stability. Exploiting low media literacy, these campaign efforts polarize the public and threaten electoral integrity, potentially amplified by external adversarial networks.

• Overall national security impact: Substantial
• Scope: Single nation
• Primary target: Kosovo
• Alleged perpetrator: Unknown

Threat characteristics

• Imminence: Near-term. The rapid and silent proliferation of deepfakes represents an active, developing situation affecting public trust and upcoming electoral processes.
• Autonomy: Human-controlled. AI systems were used as tools by human actors to generate synthetic media, with no independent decision-making or autonomous action.
• Novelty: Evolved capability. Represents an advanced, scaled deployment of generative AI tools targeting a specific, vulnerable political ecosystem, building on established deepfake methods.

Impact by dimension

• Physical security: Negligible. No physical security threats, kinetic attacks, or critical infrastructure compromise were reported in relation to this incident.
• Information security: Substantial. AI-generated deepfakes targeted top political leaders at scale on social media, with concerns of potential involvement by Kremlin-linked networks to spread disinformation.
• Sovereignty: Substantial. The proliferation of deepfakes of high-ranking officials threatens to distort electoral processes and undermine public trust in core democratic institutions.
• Economic security: Negligible. No significant economic impacts, financial system attacks, or strategic technology theft were indicated in the report.
• Societal stability: Substantial. The widespread dissemination of manipulated content exploits low media literacy, deepens social polarization, and degrades public trust in shared reality.