Purportedly AI-Generated Woman in Videos Reportedly Led Brazilian Bricklayer to Travel Over 2,000 KM and Become Stranded in Belo Horizonte, Brazil

A Brazilian bricklayer reportedly traveled over 2,000 km from Imperatriz, Maranhão, to Belo Horizonte, Minas Gerais, after believing videos of a purportedly AI-generated woman offering a relationship and life on a farm. He exhausted his money on lodging and needed municipal help to return home. The profile operator later denied having private conversations with him. The identity of the operator and the generation tool were not established.

Risk classification

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

The incident involves the use of an AI-generated persona to run romance scams, manipulating victims into believing they are interacting with a real person to gain personal or financial advantage.

Additional risk subdomains

• 5.1 Overreliance and unsafe use: The victim formed an intense emotional attachment to the AI persona due to loneliness, leading to unsafe real-world actions like traveling 2,000 km.

Causal factors

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

The incident was caused by an AI-generated persona sending automated responses to users post-deployment, intentionally designed to simulate a real person and attract interactions.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Risk Level 3: The system is an AI-generated persona and chatbot ('figura gerada por inteligencia artificial que enviava respostas automaticas') which interacts with users without disclosing its artificial nature, falling under the category of chatbots and deepfakes.

AI system and alleged parties

• AI system: None named
• AI purpose: Chatbot; Social Media Content Generation
• Behaviour type: Autonomous
• Alleged developer: Synthetic video generation technology developers, Synthetic media generation technology developers
• Alleged deployer: Synthetic media creators, Scammers in Brazil, Scammers, Romance scammers
• Alleged harmed parties: Victims of romance scams in Brazil, Victims of romance scams, Social media users, Men and boys, Men, Epistemic integrity, Edimar Rocha Martins, Construction workers

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

Malicious content

Reported: The report explicitly describes toxic or malicious content created or spread directly by the incident.

Directly caused: The AI system generated and sent automated deceptive messages and videos of attractive women to lure victims.

Psychological

Reported: The report explicitly describes psychological distress and trauma caused directly by the incident.

Directly caused: Edimar suffered emotional distress, loneliness, and trauma upon discovering the woman was an illusion ('diante do trauma de descobrir que a mulher prometida era uma ilusao digital').

Indirectly caused: The emotional toll of the deception led him to radically cut contact with all social media.

Inferred additional harm: The other two elderly victims mentioned in the report likely suffered similar psychological trauma and distress from being deceived.

People affected

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

Potential causes

Management

• Negligent Platform Oversight: Platform management failed to curb systemic automated fraud accounts.

Technology

• AI Persona Generation Tools: AI generated realistic videos and profiles to deceive users.
• Automated Response Bots: Bots sent automated messages to simulate genuine human interaction.

Data Inputs

• Deceptive Video Training Data: AI models used attractive female videos to create realistic fakes.

Human Factors

• Extreme User Loneliness: Grief and isolation made the victim highly susceptible to deception.
• Limited Digital Literacy: Victim could not distinguish between a real person and an AI bot.

Process and Methods

• Weak Profile Verification: Social media platforms allowed unverified AI bots to contact users.
• Poor Automated Detection: Lack of active monitoring for repetitive AI-driven catfishing scams.

Regulatory Environment

• Absence of AI Identity Laws: Lack of regulations requiring disclosure of AI-generated personas.

Information quality

• Classification confidence: High
• Reason for confidence: The report clearly details the victim's experience, the location, the dates, and the mechanism of deception (AI-generated persona with automated responses). It also mentions two other similar cases, providing good context. However, technical details about the specific AI model or its creators are absent.
• Ambiguities identified: The exact platform where the AI persona operated is not specified, and the identity/motives of the creators of the AI persona are unknown.
• Alternative interpretations: The incident could be viewed purely as a traditional catfishing scam where AI was merely a tool, rather than an AI-specific safety failure.

An AI-generated social media persona was used to conduct automated romance scams in Brazil, leading to emotional distress and financial loss for multiple victims. While demonstrating an evolved capability in automated fraud, the national security impact remains minor and localized, manageable by standard municipal and law enforcement services.

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

Threat characteristics

• Imminence: Long-term. Represents an ongoing, low-level societal challenge regarding automated fraud rather than an immediate national security crisis.
• Autonomy: Full autonomy. The AI persona operated independently, generating and sending automated responses to users without active human intervention in daily interactions.
• Novelty: Evolved capability. Represents a technological evolution of traditional romance scams, leveraging automated generative AI to scale emotional manipulation.

Impact by dimension

• Physical security: Negligible. No physical infrastructure, critical systems, or public safety mechanisms were targeted or affected by this localized romance scam.
• Information security: Negligible. The incident involves localized automated social media deception for personal fraud rather than coordinated information warfare or intelligence compromise.
• Sovereignty: Negligible. No impact on state authority, government decision-making, border control, or constitutional processes.
• Economic security: Negligible. Financial loss was limited to the victim's personal savings and local municipal assistance, posing no threat to national economic or technological security.
• Societal stability: Minor. Minor impact involving the targeted exploitation of vulnerable citizens, requiring assistance from local municipal migrant support services.