Purported AI-Generated Deepfake of Irish Fine Gael Presidential Candidate Heather Humphreys Used in Fake Investment Videos on Meta Platforms

Purported deepfake videos reportedly circulated on Meta platforms cloning Irish Fine Gael presidential candidate Heather Humphreys to allegedly portray her endorsing high-return investment schemes. The purportedly AI-generated image and voice cloning aimed to exploit public trust and lure consumers into fraud. The Bank of Ireland warned of further scams and risks.

Risk classification

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

The incident involves fraudsters using AI voice and image cloning to impersonate trusted public figures in order to execute financial investment scams targeting individual consumers.

Causal factors

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

The incident is caused by human fraudsters intentionally using deployed AI voice-cloning and deepfake tools to deceive victims for financial gain.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The report describes AI-generated content such as deepfakes, which fall under the transparency obligations of Risk Level 3.

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 impersonating Heather Humphreys, Unknown scammers, Quantum AI scammers, Quantum AI
• Alleged harmed parties: Heather Humphreys, General public of Ireland, Fine Gael, Epistemic integrity, Electoral 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 Minor, 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

Financial

Reported: Yes, the report explicitly describes significant financial losses, noting that Europeans lose over €4 billion annually to online financial scams, with specific victims losing between €250 and €170,000.

Directly caused: Direct losses include €15,000 lost by an Italian banker, €6,000 by an Irish engineer, and €170,000 by a German citizen, averaging tens of thousands of euros per detailed victim.

Inferred additional harm: Given the €4 billion annual loss across Europe, it is likely that thousands of other unnamed individuals have suffered similar devastating financial losses.

Malicious content

Reported: Yes, the report describes the creation and spread of malicious deepfake videos and fake ads on social media platforms.

Directly caused: Fraudsters created deepfake videos cloning the voices and images of politicians like Heather Humphreys and Pistorius to promote scams.

Inferred additional harm: It is likely that thousands of variations of these deepfake videos continue to be generated and spread across social media platforms to evade detection.

Epistemic

Reported: Yes, the report describes the fabrication of video and audio evidence to deceive users.

Directly caused: AI-driven voice and image cloning was used to fabricate endorsements from public figures, creating highly convincing fake videos.

Inferred additional harm: The proliferation of highly convincing deepfakes contributes to a broader erosion of shared reality and trust in digital media.

People affected

• Occurrences reported: 1
• People reportedly harmed: 4
• People reportedly exposed: 12300000

Potential causes

Management

• Prioritizing Revenue Over Safety: Platforms allow paid scam ads to run while blocking private posts immediately.
• Inadequate Moderation Resources: Platforms rely on chatbots and understaffed human review teams for safety.
• Ending Fact-Checking Systems: Meta ended its third-party fact-checking system, reducing oversight.

Technology

• Advancing Voice-Cloning Software: AI software easily clones voices to create highly convincing fake endorsements.
• Evasion of Automated Filters: AI-generated ads are slightly altered to bypass automated platform detection.
• Algorithmic Recommendation Loop: Platform algorithms feed more scam content to users after an initial click.

Data Inputs

• Stolen Celebrity Likeness Data: Unauthorized use of public figures' voice and image data to train deepfakes.
• Fabricated Trading Screen Data: Scam platforms display false profits to convince victims to invest more.

Human Factors

• Trust in Well-Known Figures: Victims trust cloned images of politicians and celebrities endorsing schemes.
• Manipulation by Fake Advisors: Social engineering by phone agents who pressure victims to invest more.

Process and Methods

• Slow Platform Takedown Process: Meta takes days or weeks to remove flagged fraudulent deepfake ads.
• Flawed Reporting Mechanisms: Reporting is managed by chatbots and limits flaggers to 20 URLs per report.
• Cloaking and Ad Deactivation: Scammers run ads briefly and use fifty-plus variants to evade detection.

Regulatory Environment

• DSA Article 8 Monitoring Exemption: No general obligation for platforms to monitor third-party published content.
• Lack of EU Advertiser Verification: EU regulations do not mandate pre-verification of financial advertisers.
• Limited Prosecution Resources: Law enforcement agencies are overwhelmed by the volume of cross-border scams.

Information quality

• Classification confidence: High
• Reason for confidence: The reports provide clear, detailed accounts of how the deepfakes are created, deployed, and the specific financial harms they cause to named individuals and the broader European public. The role of the AI system (voice and image cloning) is explicitly stated, and there are no conflicting accounts.

Fraudsters are utilizing advanced AI voice and image cloning to impersonate public figures and politicians in highly convincing deepfake ads, leading to substantial economic losses of over 4 billion euros annually across Europe. While primarily a cybercrime issue rather than a direct state-sponsored national security threat, the scale of economic impact and impersonation of government officials warrants notable attention from security and regulatory agencies.

• Overall national security impact: Substantial
• Scope: Multiple nations
• Primary target: European countries
• Alleged perpetrator: Unknown

Threat characteristics

• Imminence: Long-term. The incident represents an ongoing capability and strategic cybercrime concern rather than an immediate, acute national security crisis.
• Autonomy: Human-controlled. The AI voice-cloning and deepfake tools are directly operated by human fraudsters who generate and distribute the content.
• Novelty: Evolved capability. While deepfakes and financial scams are established threats, the scale, sophistication, and use of voice-cloning to evade detection represent an evolved capability.

Impact by dimension

• Physical security: Negligible. No physical security threats, kinetic attacks, or critical infrastructure disruptions were reported in connection with this incident.
• Information security: Minor. While the deepfakes impersonate high-level political figures, the primary intent is financial fraud rather than state-sponsored information warfare or intelligence compromise.
• Sovereignty: Minor. Impersonation of government ministers risks eroding public trust in official communications, but does not disrupt core government decision-making or constitutional processes.
• Economic security: Substantial. The scale of financial loss is substantial, with Europeans losing over 4 billion euros annually to online financial scams, representing a notable threat to consumer economic security.
• Societal stability: Minor. Widespread exposure to deepfakes contributes to epistemic erosion and trust issues in digital media, but does not actively trigger civil unrest or systemic human rights abuses.