Alleged Deepfake Investment Scam in Spain Defrauds 208 Victims of €19 million ($20.9 million)

Spanish police arrested six individuals allegedly behind a €19M ($20.9M) global investment scam powered by AI. The operation used deepfake ads featuring national celebrities to deceive victims, many of whom were selected through targeting algorithms. Scammers posed as financial advisors and fake officials, cycling through romance baiting, investment fraud, and recovery scams. AI-generated content amplified trust and engagement.

Risk classification

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

The incident involves scammers using generative AI tools to create highly realistic deepfakes, fake passports, and cloned websites to execute targeted financial fraud and investment scams.

Additional risk subdomains

• 3.1 False or misleading information: AI-generated deepfakes and cloned trading platforms were used to spread false information and mislead victims into believing the investments were legitimate.

Causal factors

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

The incident was caused by human scammers intentionally using deployed generative AI tools to create fake passports, deepfakes, and cloned websites to defraud victims.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The report describes the use of 'AI-generated content such as deepfakes' to create fake ads of public figures and fake passport images.

AI system and alleged parties

• AI system: ChatGPT, unspecified (OpenAI)
• AI purpose: Deepfake Video Generation; Deepfake Image Generation
• Behaviour type: Tool
• Alleged developer: Unknown generative AI developers, Unknown deepfake technology developers
• Alleged deployer: Unnamed criminal group in Spain, Individuals arrested in Granada and Alicante
• Alleged harmed parties: Individuals who lost money through scam investment platforms, Global victims targeted by scams, General public of Spain, General public

Harm severity

Severity ratings are based on this scale.

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

Financial

Reported: The report describes multiple operations causing total financial losses of approximately 303.2 million dollars across 4 occurrences, with an average loss of 75.8 million dollars per occurrence.

Directly caused: Scammers directly stole 35.3 million dollars from Georgian victims, 247 million dollars from European victims, and 20.9 million dollars from Spanish victims, averaging 75.8 million dollars per occurrence.

Indirectly caused: Victims incurred indirect costs such as bank fees, legal fees, and costs trying to recover funds.

Inferred additional harm: Total global financial losses from these AI-enabled scams are likely much higher, potentially exceeding 500 million dollars, as many scams go unreported.

Malicious content

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

Directly caused: Scammers created and spread AI-generated deepfake ads featuring popular public figures on social networks to lure victims.

Inferred additional harm: Widespread dissemination of deepfake videos and fake passport images across various online platforms and social media networks, exposing millions of users to malicious content.

Differential treatment

Reported: The report explicitly describes differential treatment of individuals caused directly by the incident.

Directly caused: Spanish police noted that 'algorithms selected those whose profiles matched the cybercriminals' searches' to target them with specific ads.

Inferred additional harm: Algorithmic profiling likely systematically targeted vulnerable demographics, such as elderly or isolated individuals, for romance and investment scams.

Privacy

Reported: The report explicitly describes privacy violations caused directly by the incident.

Directly caused: Scammers used algorithms to search and select victims based on their online profiles and personal data.

Inferred additional harm: Unauthorized harvesting and processing of personal data from social media platforms to profile and target potential victims.

Psychological

Reported: The report explicitly describes psychological harm, noting that scammers simulated romantic relationships ('romance baiting') to manipulate victims.

Directly caused: Scammers simulated romantic relationships and posed as financial advisors, causing emotional distress and manipulation to at least 208 victims in the Spanish operation.

Indirectly caused: The loss of life savings and trust from highly sophisticated scams likely caused severe anxiety, distress, and trauma to the 33,209 victims.

Inferred additional harm: Widespread psychological trauma, anxiety, and depression among the tens of thousands of victims who lost significant financial assets.

Epistemic

Reported: The report explicitly describes epistemic harm caused directly by the incident.

Directly caused: Scammers used AI to generate deepfake videos of public figures and cloned websites (Exante, Admirals) to fabricate a false reality of legitimate investment opportunities.

Indirectly caused: The proliferation of highly realistic deepfakes erodes public trust in video evidence and legitimate financial institutions.

Inferred additional harm: Widespread erosion of trust in digital media, online advertisements, and financial platforms due to the difficulty of distinguishing real content from AI-generated fakes.

People affected

• Occurrences reported: 4
• People reportedly harmed: 33209
• People reportedly exposed: 100000

Potential causes

Management

• Inadequate Bank Verification: JP Morgan Chase allowed scammers to open real accounts with fake identities.
• Brokerage Lack of US Presence: Exante could not easily report or stop scams targeting US clients.

Technology

• AI-Generated Deepfake Video Ads: Scammers used AI to generate highly convincing deepfake ads of celebrities.
• Rapid AI Website Cloning Tools: AI tools allowed scammers to quickly clone complex trading boards and sites.
• ChatGPT Fake Passport Generation: ChatGPT image generation made creating realistic fake passports extremely easy.

Data Inputs

• Victim Profiling Algorithms: Algorithms analyzed user profiles to select and target vulnerable victims.
• Real Platform Data Replication: Scammers replicated real trading board data to make fake platforms look real.

Human Factors

• Trust in Deepfake Celebrity Ads: Victims trusted fake ads due to realistic deepfakes of public figures.
• Susceptibility to Romance Baiting: Victims fell for emotional manipulation and fake romantic interest by scammers.
• Overreliance on Fake Digital Figures: Victims trusted fake platform interfaces showing fabricated high returns.

Process and Methods

• Fast Creation and Exploitation: AI-driven automation drastically reduced the scammers' time-to-exploit.
• Multi-Phase Scam Tactics: Scammers used complex multi-phase tactics, including fake recovery claims.
• Shell Company Money Laundering: Criminals set up multiple shell companies and aliases to launder proceeds.

Regulatory Environment

• Lack of Active AI Fraud Oversight: Regulators have not fully explored AI capabilities to address AI-driven fraud.
• Low Barriers to Entry for Actors: Lack of strict controls allows cybercriminals easy access to advanced AI tools.

Information quality

• Classification confidence: High
• Reason for confidence: The reports provide detailed, corroborated accounts from financial compliance officers, regulators (CFTC, SEC, FBI), and law enforcement (Spanish police) regarding the scale, methods, and financial impact of the scams. The role of AI in generating deepfakes, fake passports, and cloned websites is clearly documented with specific figures on victims and losses.
• Ambiguities identified: The specific AI models used for deepfake video generation and website cloning are not named, though ChatGPT is mentioned for passport generation.
• Alternative interpretations: None. The events are clearly identified as coordinated financial fraud facilitated by generative AI tools.

Criminal organizations utilized generative AI tools, including deepfakes and synthetic document generation, to conduct large-scale international financial fraud. While the operations caused substantial economic harm to tens of thousands of individuals globally, the overall national security impact remains minor as the incident did not target critical infrastructure, sovereignty, or systemic stability.

• Overall national security impact: Minor
• Scope: Multiple nations
• Primary target: No clear primary
• Other affected: Spain, Georgia, Israel, United States, Bulgaria, Cyprus
• Alleged perpetrator: Criminal organizations

Threat characteristics

• Imminence: Long-term. The threat represents an ongoing strategic concern regarding the scaling of AI-enabled cybercrime rather than an imminent national security crisis.
• Autonomy: Human-controlled. AI tools were utilized as assistive instruments to generate specific content, while human operators maintained full control over the fraudulent schemes.
• Novelty: Evolved capability. The incident demonstrates a significant evolution of traditional financial fraud through the integration of generative AI to scale and automate deceptive content.

Impact by dimension

• Physical security: Negligible. No physical infrastructure, kinetic targeting, or critical safety systems were affected by these financial scams.
• Information security: Minor. Deepfakes of public figures were deployed on social media, but the primary intent was financial fraud rather than state-sponsored information warfare or intelligence compromise.
• Sovereignty: Negligible. Core government operations, electoral systems, and sovereign decision-making processes were not targeted or compromised.
• Economic security: Minor. The operations resulted in substantial financial losses of over 300 million dollars for retail investors, though they did not threaten systemic national financial infrastructure.
• Societal stability: Minor. Tens of thousands of victims experienced psychological and financial harm, eroding trust in digital media, but the incident did not cause widespread civil unrest.