This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A 62-year-old woman in New Taipei City was defrauded of NT$2.64 million by scammers who used AI face-changing and voice-cloning technology to impersonate a Hong Kong entertainer during video calls. The victim, a long-time fan, was convinced to wire money for a purported visit to Taiwan. Police intervened after the victim's family became suspicious, leading to the arrest of a woman who had been recruited by the same scam group to collect cash.
Scammers allegedly defrauded a woman in New Taipei City of NT$2.64 million (US$81,116) by impersonating Hong Kong entertainer Andy Lau using purported deepfakes. The alleged scam convinced the victim, a long-time fan, through a video call that "Lau" needed funds for a visit to Taiwan. The victim reportedly wired the money, but her family suspected a scam and involved the police. An alleged scammer was arrested after attempting to collect a staged cash payment.
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 scammers used AI face-changing and voice-cloning technology to impersonate a celebrity and target a specific victim for financial fraud.
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 incident was caused by human scammers who intentionally deployed and used AI face-changing and voice-cloning tools to defraud a victim.
Added categories peak at 2: Minor (Psychological Harm, Inferred)
A localized financial fraud incident in Taiwan where scammers used real-time AI face-changing and voice-cloning to impersonate a celebrity and defraud an individual of NT$2.64 million. While demonstrating the evolving accessibility of deepfake tools for criminal activity, the incident is a standard law enforcement matter with negligible national security implications.
The AI face-changing and voice-cloning applications were used as tools by human scammers to generate specific synthetic media (video and voice) based on direct inputs to impersonate a celebrity.
Limited Risk: The AI systems used are deepfakes (AI-generated content), which under the EU AI Act pose a moderate risk and are subject to transparency obligations to inform users they are interacting with AI-generated content.
The scammers used AI face-changing apps to generate a deepfake video of Andy Lau during a video call, and AI voice generation to clone his voice.
The reports from TVBS and Taipei Times provide consistent, clear details about the scam, the specific amount stolen (NT$2.64 million), the technology used (AI face-changing and voice-cloning), the arrest date, and the individuals involved. There is very little ambiguity regarding the events.
None
Technology
AI Face-Changing Technology
Deepfake technology allowed scammers to realistically mimic the star's face.
AI Voice Generation
AI was used to clone the entertainer's voice, increasing credibility.
Real-Time Video Manipulation
Enabled live video interaction, convincing the victim the call was genuine.
Data Inputs
Celebrity Media Data
Abundant public images and videos of the star were used to train AI models.
Human Factors
Starstruck Vulnerability
The victim's deep admiration for the celebrity clouded her judgment.
Misplaced Trust in Video
Victim naively believed video calls were foolproof proof of identity.
Deception of Money Mule
Another fan was manipulated into acting as a courier under false pretenses.
Process and Methods
Social Media Impersonation
Scammers initiated contact through fake celebrity fan groups on Facebook.
Urgent Financial Requests
Fabricated stories of travel expenses pressured the victim to wire money.
Management
Inadequate Platform Moderation
Social media platforms failed to detect and remove fake fan pages.
Regulatory Environment
Lack of Deepfake Regulation
Absence of strict laws governing the creation and use of AI-cloning apps.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
