
The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.
A man in Taiwan used AI deepfake technology to create a forged passport photo by blending his and another person's facial features, attempting to evade prison by illegally crossing the border at Kinmen. The AI-powered e-Gate facial recognition system detected the anomaly, leading to his arrest and prosecution for identity fraud and document forgery.[AI generated]
Why's our monitor labelling this an incident or hazard?
The event explicitly involves the use of an AI system (AI deepfake technology) to create a fraudulent identity document photo, which directly led to an illegal act (identity fraud) and attempted border crossing. The AI system's use is central to the harm (violation of legal frameworks and potential security risks). The detection system (also AI-based) prevented the harm from fully materializing but the incident of misuse and attempted fraud has occurred. Hence, it meets the criteria for an AI Incident due to the realized harm and legal violations stemming from AI misuse.[AI generated]