Family codewords offer simple defence against growing AI…

Elizabeth Benz was targeted in a sophisticated deepfake scam involving a caller who successfully imitated the voice of her 16-year-old son, John. The experience, which occurred on August 9, 2026, left Benz overwhelmed by a sense of extreme terror as the caller claimed her son was involved in a fatal accident and subsequently threatened his life.

What happened

The attack utilized AI-generated voice synthesis to create a convincing emergency scenario designed to bypass the victim's critical thinking. By simulating her son sobbing and claiming he was in immediate danger, the scammers aimed to induce a state of panic that would prevent Benz from verifying the caller's identity. Although Benz had a family safeguard in place, she found that the intense emotional pressure of the situation made it impossible to recall or utilize the precaution at the time. The scammers leveraged psychological manipulation, including threats of violence, to control the narrative and urgency of the call. Benz later discovered her son was safe and unharmed, and she has since shared her story to raise awareness about the psychological realism of AI-enabled fraud. Experts note that these scams often exploit a sense of urgency, demands for untraceable payments, and efforts to isolate the victim from seeking outside help. Security professionals now recommend establishing a private family codeword to verify identity during unusual or high-stress requests, emphasizing that preparation and mental training are essential to overcoming the panic these realistic deepfakes are designed to trigger.