Purportedly AI-Generated Facebook Video Allegedly Misused Skënder Brataj and Blendi Fevziu to Promote Purported Miracle Cream in Albania

A purportedly manipulated AI-generated video reportedly circulated on Facebook misusing the likenesses of emergency physician Skënder Brataj and journalist Blendi Fevziu to promote a purported "miracle" cream. Reporting indicated that some citizens believed the content and contacted or visited emergency services seeking treatment.

Risk classification

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

The incident involves the use of AI-generated deepfakes of public figures to deceive citizens for financial gain through a fraudulent product scam, matching boundary rule B1.

Additional risk subdomains

• 3.1 False or misleading information: The deepfake video spreads false medical and news information, leading to inaccurate beliefs and unsafe real-world actions by patients.

Causal factors

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

The incident was caused by malicious human actors who intentionally created and deployed a deepfake video to scam the public.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The report describes the use of AI-generated deepfakes, which are subject to transparency obligations under the EU AI Act.

AI system and alleged parties

• AI system: None named
• AI purpose: Deepfake Video Generation; Voice Generation
• Behaviour type: Tool
• Alleged developer: Unknown voice cloning technology developers, Unknown deepfake technology developers
• Alleged deployer: Unknown scammers
• Alleged harmed parties: Skënder Brataj, Public trust in medical guidance, People seeking medical advice, General public of Albania, General public, Epistemic integrity, Blendi Fevziu

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 Negligible
• Privacy: direct Negligible, indirect Negligible
• Psychological: direct Negligible, indirect Minor
• Epistemic: direct Minor, indirect Minor
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Malicious content

Reported: The report explicitly describes a manipulated deepfake video circulating on Facebook.

Directly caused: An AI-generated deepfake video was created and spread on Facebook, falsely claiming a physical attack on a doctor by a 'pharmaceutical mafia'.

Inferred additional harm: The malicious video likely reached thousands of Facebook users in Albania before or despite being reported.

Psychological

Reported: The report explicitly describes worry and concern from the impersonated individuals and the public.

Indirectly caused: Dr. Skender Brataj and journalist Blendi Fevziu experienced distress and concern over the misuse of their likenesses. Dr. Brataj also responds daily to anxious and worried citizens.

Inferred additional harm: Dozens of elderly or anxious patients likely experienced confusion and distress regarding their health and the fabricated attack on Dr. Brataj.

Epistemic

Reported: The report explicitly describes a fabricated news report and false medical claims.

Directly caused: The video fabricated a news broadcast by Blendi Fevziu and a medical endorsement by Dr. Brataj, eroding the shared information environment.

Indirectly caused: The fabrication led citizens to believe false claims about a physical attack and a miracle cure.

Inferred additional harm: The ongoing circulation of such deepfakes erodes public trust in legitimate media and public health authorities in Albania.

People affected

• Occurrences reported: 1
• People reportedly harmed: 3
• People reportedly exposed: 1000

Potential causes

Management

• Inadequate Moderation Resources: Platform management underinvested in local-language content moderation.

Technology

• AI Deepfake Tools: AI tools allowed realistic cloning of faces and voices of public figures.
• Algorithmic Distribution: Social media algorithms promoted the viral spread of the deepfake video.

Data Inputs

• Misused Public Media Data: Publicly available broadcasts were used to train or feed the AI model.

Human Factors

• Public Trust in Media Figures: Victims trusted the familiar doctor and journalist, leading to deception.
• Lack of Digital Literacy: Vulnerable individuals struggled to distinguish deepfakes from real news.

Process and Methods

• Slow Platform Takedown Process: Reporting mechanisms failed to quickly remove the reported scam videos.
• Inadequate Identity Verification: Lack of robust checks allowed fake accounts to publish deceptive ads.

Regulatory Environment

• Cross-Border Legal Gaps: Perpetrators outside Albania evaded local cybercrime law enforcement.
• Weak AI Safety Regulations: Lack of strict legal frameworks targeting AI-generated scam content.

Information quality

• Classification confidence: High
• Reason for confidence: The report clearly details the nature of the deepfake, the individuals impersonated, the platform used, and the specific real-world consequences of the scam.
• Ambiguities identified: The specific AI tools used to generate the deepfake are not named.
• Alternative interpretations: None, the incident is a straightforward deepfake-enabled scam.

The incident involves a deepfake-enabled medical scam in Albania that impersonated the national emergency chief and a prominent journalist. While it caused minor real-world disruptions to public health services and eroded public trust, it remains a localized cybercrime issue rather than a significant national security threat.

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

Threat characteristics

• Imminence: Near-term. The scam continues to circulate and impact public health resources, requiring ongoing monitoring and cybercrime response.
• Autonomy: Human-controlled. The AI was used purely as a tool by human creators to generate synthetic video and voice cloning assets.
• Novelty: Established threat. Deepfake-enabled medical and financial scams targeting public figures are a well-established global threat.

Impact by dimension

• Physical security: Minor. Minor impact on healthcare services as patients visited the national emergency center seeking fake treatment, but no critical infrastructure compromise.
• Information security: Minor. Use of deepfakes to impersonate a journalist and national emergency chief, but aimed at financial fraud rather than state-sponsored intelligence or information warfare.
• Sovereignty: Minor. Impersonation of the national emergency chief impacted public health communications, but did not disrupt core government functions or state sovereignty.
• Economic security: Negligible. No strategic economic or technological threat; impact is limited to localized consumer fraud.
• Societal stability: Minor. Erodes public trust in national health authorities and media, but does not threaten broader social cohesion or human rights.