Alleged Deepfake of Luis Suárez Used to Scam Uruguayans in Fake Investment Scheme

On December 26, 2024, UTE, Uruguay's public utility company, warned of an alleged deepfake video circulating on Instagram, Facebook, and Threads. The video reportedly features manipulated images and voice of footballer Luis Suárez, falsely promoting an investment scheme claiming a 10,400 peso investment could yield 175,000 pesos monthly. It also reportedly uses AI-altered Telemundo footage. UTE urged the public not to share personal data, as the scam links to phishing sites posing as official sources.

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 to impersonate a public utility company and a famous footballer to execute a targeted financial scam and phishing campaign.

Causal factors

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

The incident was caused by human actors who intentionally deployed AI tools to generate deepfakes for a financial scam after the AI models were trained and deployed.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Risk Level 3. Limited Risk: The report describes the creation of AI-generated deepfakes. According to the definitions, 'AI-generated content such as deepfakes' falls under Limited Risk, which requires notifying users when they are interacting with such content.

AI system and alleged parties

• AI system: unspecified
• AI purpose: Deepfake Video Generation; Voice Generation
• Behaviour type: Tool
• Alleged developer: Unknown deepfake technology developers
• Alleged deployer: Unknown scammers, Braziko
• Alleged harmed parties: UTE, Uruguayan general public, Telemundo, Luis Suárez, El País (Spain)

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

Malicious content

Reported: Yes, the report describes a malicious deepfake video and fake news article used to scam users.

Directly caused: The scammers created and distributed a manipulated video altering the voice and image of a famous footballer and a news reporter to promote a fake investment.

Inferred additional harm: The malicious content was likely viewed by thousands of social media users on Facebook and Instagram before being taken down.

Epistemic

Reported: Yes, the report describes the fabrication of a news broadcast and a newspaper article using AI.

Directly caused: A Telemundo news report was digitally altered using AI to show a presenter and a footballer endorsing a fake investment scheme.

Inferred additional harm: This contributes to the erosion of trust in public figures, news organizations, and official communications from public utilities like UTE.

People affected

• Occurrences reported: 1
• People reportedly exposed: 10000

Potential causes

Management

• Inadequate Proactive Monitoring: UTE relied on reactive warnings instead of proactive threat intelligence.
• Delayed Ad Takedown Response: Social media platforms were slow to remove active scam advertisements.

Technology

• AI Deepfake Generation Tools: AI was used to alter the voice and image of a soccer player and reporter.
• Social Media Ad Delivery: Meta platforms allowed rapid distribution of fraudulent short-duration ads.
• Malicious Domain Registration: Domain registrars permitted quick setup of deceptive phishing domains.

Data Inputs

• Unauthorized Media Harvesting: Scammers used existing videos of the reporter and player to train AI models.
• Stolen Corporate Branding: Official logos of UTE and ministries were used to make fake sites look real.

Human Factors

• User Susceptibility to Scams: Financial vulnerability makes users trust high-return investment promises.
• Trust in Public Figures: Users trusted the deepfaked soccer player and recognized news reporter.

Process and Methods

• Weak Ad Review Processes: Meta platforms failed to detect and block deepfake scam ads before airing.
• Lack of Domain Verification: No verification process prevented the registration of 'uteinvst.online'.

Regulatory Environment

• Lack of Deepfake Regulations: Insufficient legal framework to prevent or penalize AI identity theft.
• Poor Cross-Border Enforcement: Difficult to prosecute actors operating from Ukraine or Latvia.

Information quality

• Classification confidence: High
• Reason for confidence: The reports clearly describe the nature of the scam, the platforms used, the specific public figures targeted, and the technical mechanism (AI-altered video and voice). While the exact AI model used is not specified, the classification of the incident as a deepfake-enabled financial scam is highly unambiguous.
• Ambiguities identified: The exact number of victims who lost money and the total financial loss are not specified.

AI-generated deepfakes of a famous footballer and news reporter were used to impersonate Uruguay's public utility (UTE) in a financial scam. While it represents an evolving threat of digital impersonation, it has minor impact on national security, critical infrastructure, or sovereignty.

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

Threat characteristics

• Imminence: Long-term. Represents an ongoing strategic concern regarding deepfake-enabled fraud rather than an active, immediate crisis.
• Autonomy: Human-controlled. AI was used as a tool to generate deepfake media under direct human control, with no autonomous decision-making.
• Novelty: Established threat. Deepfakes used for financial scams and phishing are an established threat with numerous prior occurrences globally.

Impact by dimension

• Physical security: Negligible. Impersonation of public utility name and logo for financial scam with no manipulation of actual critical infrastructure or control systems.
• Information security: Minor. AI-generated deepfakes on social media platforms used for financial fraud rather than coordinated state-sponsored information warfare or intelligence compromise.
• Sovereignty: Minor. Impersonated public utility and used government logos on a phishing site, but did not disrupt core government decision-making or sovereign operations.
• Economic security: Negligible. Individual-level financial fraud and phishing. No systemic threat to Uruguay's financial systems or strategic technologies.
• Societal stability: Minor. Erosion of trust in public utility and media through deepfakes, but lacks large-scale impact on civil liberties or societal stability.