Video with Reportedly AI-Generated Media Purported to Show Croatian Neurosurgeon Josip Paladino Endorsing Steplex in Fake TV Segment on Facebook

A Facebook page reportedly posted a video styled as a TV news segment that purported to show Croatian neurosurgeon Josip Paladino endorsing a pain-treatment product called Steplex after an on-air assault. The clip was reported to use AI-generated or manipulated media built from older footage, and the ad linked users to a page mimicking Croatia's Ministry of Health while soliciting purchases and contact data.

Risk classification

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

The AI system was used to create deepfake videos impersonating public figures to deceive users for financial gain through a medical scam.

Additional risk subdomains

• 3.1 False or misleading information: The deepfake generated and spread false claims about an unapproved medical product, potentially misleading users about health treatments.

Causal factors

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

The incident was caused by malicious human actors who intentionally generated and deployed deepfake videos to perpetrate a financial scam.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

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

AI system and alleged parties

• AI system: unspecified
• 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, Green Facebook page
• Alleged harmed parties: People seeking medical advice, Ministry of Health (Croatia), Ministarstvo zdravstva, Josip Paladino, General public of Croatia, Epistemic integrity

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

Malicious content

Reported: The report explicitly describes malicious deepfake content created and spread on Facebook.

Directly caused: A deepfake video impersonating Sasa Kopljar and Josip Paladino was created and posted on a Facebook page named Green.

Privacy

Reported: The report explicitly describes the collection of personal data on the fraudulent website.

Indirectly caused: The scam website set up via the deepfake campaign solicited personal data, specifically names and phone numbers, from visitors.

Inferred additional harm: It is inferred that users who submitted their details had their personally identifiable information compromised and potentially exploited for further scams.

Epistemic

Reported: The report explicitly describes epistemic harm through the fabrication of a news segment and an exclusive interview.

Directly caused: The deepfake video fabricated a news segment claiming a physical fight occurred in a studio and that Dr. Paladino endorsed an unapproved drug.

Indirectly caused: A fraudulent website mimicking the Croatian Ministry of Health was created to host false clinical claims and sell the fake drug.

Inferred additional harm: The incident contributes to the erosion of public trust in legitimate medical professionals, health authorities, and broadcast journalism.

People affected

• Occurrences reported: 1
• People reportedly harmed: 2
• People reportedly exposed: 2

Potential causes

Management

• Inadequate Platform Vetting: Ad platforms did not verify identities of new pages running ads.

Technology

• AI Voice Synthesis: Allowed replication of public figures' voices with realistic tone.
• AI Video Lip-Syncing: Altered original video lip movements to match the fake audio.

Data Inputs

• Public Video Archives: Abundant YouTube videos of victims provided training data for AI.
• Spliced Historical Footage: Old unrelated videos were used to fabricate the visual narrative.

Human Factors

• Trust in Public Figures: Exploited the reputation of a trusted doctor and news anchor.

Process and Methods

• Flawed Ad Moderation: Facebook moderation failed to flag and block deepfake video ads.
• Deceptive Website Cloning: Cloned official health ministry pages to validate the AI claim.

Regulatory Environment

• Weak Enforcement: Prior police reports failed to stop recurring deepfake fraud.

Information quality

• Classification confidence: High
• Reason for confidence: The fact-checking report from Faktograf.hr provides clear, verified details about the deepfake video, the source footage used, the fraudulent website, and the unapproved status of the drug. The roles of the AI and the human actors are unambiguous.
• Ambiguities identified: The specific AI software or platform used to generate the deepfake is not identified in the report.
• Alternative interpretations: None. The evidence clearly points to a coordinated deepfake-enabled financial scam.

A deepfake video campaign impersonated a Croatian neurosurgeon and news anchor to promote an unapproved medical product, directing users to a fraudulent website mimicking the Croatian Ministry of Health. The incident represents a localized financial scam with minor national security implications, manageable under standard law enforcement and cybercrime procedures.

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

Threat characteristics

• Imminence: Long-term. Represents an ongoing strategic concern regarding the proliferation of commercial deepfakes rather than an active national security crisis.
• Autonomy: Human-controlled. AI tools were used strictly as assistance to generate the synthetic media, with humans controlling the creation and distribution.
• Novelty: Established threat. Deepfakes used for financial fraud and impersonating public figures are well-established methods of cybercrime.

Impact by dimension

• Physical security: Negligible. No physical threats, kinetic attacks, or critical infrastructure compromises occurred. The incident involves a digital medical scam.
• Information security: Minor. Involves deepfake technology to impersonate a news anchor and doctor, but constitutes a localized financial scam rather than a coordinated state-sponsored information warfare campaign.
• Sovereignty: Minor. The fraudulent website mimicked the Croatian Ministry of Health, representing a minor abuse of government authority branding, manageable under standard cybercrime procedures.
• Economic security: Negligible. Minimal economic impact restricted to individual scam victims, with no threat to strategic industries, financial systems, or technological competitiveness.
• Societal stability: Negligible. No threats to societal stability, civil liberties, or mass surveillance. The impact is limited to localized consumer fraud.