This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A woman was defrauded of over NZ$375,000 over a two-year period by a scammer posing as a celebrity. The scammer utilized AI-generated voice messages to build emotional trust and manipulate the victim into sending money for various fictitious emergencies. The victim discovered the fraud after realizing the actor's public activities contradicted the scammer's claims.
An alleged online romance scam exploiting AI-assisted impersonation defrauded a woman, "Lea," of NZ$375,000 over two years. The scammer allegedly used AI-generated voice messages and deceptive text communication to pose as New Zealand actor Martin Henderson and is reported to have built up a relationship beginning in early 2023. The victim, believing the scam was real, reportedly relocated to New Zealand in December 2024, only discovering the fraud in February 2025.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
Using AI systems to gain a personal advantage over others such as through cheating, fraud, scams, blackmail or targeted manipulation of beliefs or behavior. Examples include AI-facilitated plagiarism for research or education, impersonating a trusted or fake individual for illegitimate financial benefit, or creating humiliating or sexual imagery.
The incident involves a targeted romance scam where a perpetrator used AI-cloned voice messages to impersonate a celebrity and defraud a specific victim of $375,000.
Human
Due to a decision or action made by humans
Intentional
Due to an expected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed
The scam was orchestrated by a human actor who intentionally used AI voice cloning technology post-deployment to defraud the victim.
Added categories peak at 3: Substantial (Psychological Harm, Inferred)
An individual was defrauded of over $375,000 in a romance scam utilizing AI-cloned voice messages. While representing a tragic personal and financial loss, the incident is a localized criminal matter with negligible direct implications for national security.
The AI system was used as a tool to generate specific voice messages mimicking the actor's voice. It performed a well-defined function of voice generation based on direct inputs without autonomous planning or context retention.
Risk Level 3: Limited Risk. The system involves AI-generated content such as deepfakes (AI-doctored voice messages), which pose a moderate risk and require transparency obligations so users are aware they are interacting with AI-generated content.
The AI system was specifically used to generate voice messages mimicking the actor's voice to deceive the victim.
The reports provide consistent and detailed accounts of the scam, the amount of money lost, the specific tactics used (including AI voice cloning), and the emotional impact on the victim. Multiple sources corroborate the key facts of the incident.
None
Technology
AI Voice Cloning Technology
Scammer used AI-doctored voice messages to mimic the actor's voice.
Unregulated Communication Platforms
WeChat and Telegram enabled unverified accounts to contact users.
Lack of Deepfake Detection Tools
No automated tools on messaging apps flagged the AI-generated voice.
Data Inputs
Celebrity Public Media Data
Public voice recordings of the actor were used to train the AI model.
Victim's Personal Information
Victim shared details of family trauma, allowing targeted manipulation.
Human Factors
Victim Vulnerability
Victim faced divorce and parents' illnesses, increasing vulnerability.
Overreliance on Voice Verification
Victim trusted the AI voice messages as proof of identity.
Cognitive Dissonance and Denial
Victim ignored official warnings, believing the scammer's excuses.
Process and Methods
Inadequate Verification Processes
Fan pages and messaging apps lacked robust identity verification.
Delayed Financial Fraud Detection
Banks took two years to detect and flag suspicious transactions.
Ineffective Platform Moderation
Social media platforms failed to proactively remove fake accounts.
AI system involved
An AI system was involved, but the reports do not identify it.
Identified by automated extraction; can differ from the alleged developers above.
