Reportedly Sustained Multi-Celebrity Deepfake Persona Scam Targeting Vulnerable Southampton Resident

Over about five months in 2025, Paul Davis, a Southampton, UK man, reports that he was repeatedly targeted by scammers using purported deepfake videos and images of celebrities including Jennifer Aniston, Mark Zuckerberg, Elon Musk, and Ellie Goulding. The perpetrators allegedly rotated personas to sustain a romance and prize scam, extracting £200 in gift cards. This case suggests a new shift from one-off celebrity deepfakes to persistent, multi-persona targeting of a single vulnerable victim.

Risk classification

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

The incident involves scammers using AI-generated deepfakes to impersonate celebrities and conduct a targeted romance and financial fraud scam for personal gain.

Causal factors

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

The harm was caused by the intentional actions of human scammers who deployed readily available AI deepfake tools to defraud the victim.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The incident involves the use of AI-generated content such as deepfakes, which are classified as limited risk under the EU AI Act and are subject to specific transparency and disclosure obligations.

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
• Alleged harmed parties: Paul Davis, Mark Zuckerberg, Jennifer Aniston, General public of United Kingdom, General public of Southampton, United Kingdom, Emotionally vulnerable individuals, Elon Musk, Ellie Goulding

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 Minor
• 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

Financial

Reported: The report explicitly describes significant financial losses resulting from the scams. Paul Davis lost £200 ($270), his friend lost over £1,000, and a French woman lost £700,000 ($850,000).

Directly caused: Paul Davis suffered a direct financial loss of £200 ($270) in non-refundable Apple gift cards. The average financial loss for his specific occurrence was £200.

Indirectly caused: A French woman lost £700,000 ($850,000) to a similar AI-facilitated romance scam, and an acquaintance of Davis lost over £1,000.

Inferred additional harm: The 34 other victims associated with the unmasked Nigerian scammer, along with countless other unreported victims of similar deepfake scams, have likely suffered millions of dollars in collective financial losses.

Malicious content

Reported: The report explicitly describes the creation and spread of malicious deepfake content used to deceive the victim.

Directly caused: Scammers generated and sent highly realistic deepfake videos and audio clips of Jennifer Aniston, Elon Musk, and Mark Zuckerberg to manipulate the victim.

Inferred additional harm: It is highly likely that scammers generate and distribute thousands of similar deceptive celebrity deepfakes daily across social media platforms to target vulnerable individuals.

Psychological

Reported: The report explicitly describes psychological and mental-health harm to the victims of the scams.

Directly caused: Paul Davis, who battles depression, reported that the relentless harassment and realization of the scam severely affected his mental health, leaving him feeling humiliated and emotionally drained.

Indirectly caused: The French woman targeted in a similar romance scam reportedly lost her marriage and her mental health due to the year-long deception.

Inferred additional harm: It is highly likely that other victims of these highly manipulative romance scams suffer from severe emotional distress, anxiety, shame, and depression.

People affected

• Occurrences reported: 1
• People reportedly harmed: 3
• People reportedly exposed: 4

Potential causes

Management

• Inadequate Platform Safeguards: Facebook failed to proactively prevent automated deepfake distribution.
• Flawed Reporting Mechanisms: Platform handles user reports piecemeal without systematic resolution.
• Neglecting Vulnerable User Risks: Failure to design safety features protecting vulnerable users from scams.

Technology

• Highly Convincing Deepfakes: AI-generated videos and cloned voices mimic celebrities with high realism.
• Accessible Generative AI Tools: Free and easy-to-use AI tools allow scammers to easily fabricate media.
• Neural Network Image Generation: AI models generate customized fake certificates and prize notifications.

Data Inputs

• Publicly Available Celebrity Media: Scammers use public photos and voice samples to train AI models.
• Victim Personal Data Harvesting: Scammers scrape names and details from social media to refine AI outputs.
• Lack of Authenticated ID Inputs: Absence of secure, verified digital identity credentials on platforms.

Human Factors

• Vulnerability Due to Depression: Victim isolation and mental health struggles made him susceptible.
• Parasocial Celebrity Bonds: Victims form one-sided emotional trust with famous public figures.
• Limited Technological Literacy: Difficulty distinguishing AI-generated content from real interactions.

Process and Methods

• Slow Platform Moderation Actions: Social media platforms are slow or ineffective at blocking fake accounts.
• Unregulated Gift Card Payments: Use of non-refundable, untraceable payment methods like Apple gift cards.
• Inadequate Verification Processes: Lack of robust real-time verification for celebrity profiles.

Regulatory Environment

• Lagging Cybercrime Legislation: Laws struggle to keep up with rapid generative AI advancements.
• Jurisdictional Enforcement Barriers: Scammers operate internationally, hindering law enforcement prosecution.
• Lack of Platform Liability Rules: Absence of strict regulations holding platforms liable for hosting scams.

Information quality

• Classification confidence: High
• Reason for confidence: The reports provide consistent, detailed accounts of the scam targeting Paul Davis, including the specific amount lost, the platforms used, and the types of AI-generated media involved. Multiple sources corroborate the core facts of the incident.
• Ambiguities identified: The specific AI tools or platforms used by the scammers to generate the deepfakes are not named.
• Alternative interpretations: None.

This incident involves the use of generative AI tools to conduct romantic and financial scams against individuals. While causing personal distress and financial loss, the national security impact remains minor as it lacks state-sponsored coordination, systemic economic threats, or target-rich intelligence manipulation.

• Overall national security impact: Minor
• Scope: Multiple nations
• Primary target: No clear primary
• Other affected: United Kingdom, France
• Alleged perpetrator: Nigerian scammer

Threat characteristics

• Imminence: Long-term. Represents an ongoing strategic concern regarding the proliferation of fraud-enabling AI tools rather than an active national security crisis.
• Autonomy: Human-controlled. The AI tools were directly operated by human scammers to generate specific content, with no autonomous decision-making by the AI.
• Novelty: Evolved capability. Represents an evolution of traditional romance and phishing scams by incorporating highly realistic generative AI media.

Impact by dimension

• Physical security: Negligible. No physical security, critical infrastructure, or kinetic threats were reported in relation to this incident.
• Information security: Minor. Deepfakes of high-profile figures were used, but for localized financial fraud rather than coordinated state-sponsored information warfare.
• Sovereignty: Negligible. No government systems, electoral processes, or sovereign functions were compromised or targeted in this incident.
• Economic security: Minor. Individual financial losses occurred, including one victim losing 700,000 pounds, but there was no systemic threat to financial systems or strategic economic security.
• Societal stability: Minor. The incident caused psychological harm and financial distress to specific individuals, but did not threaten large-scale societal stability or civil liberties.