In January 2023, Arizona resident Jennifer DeStefano picked up a call from an unknown number, assuming it was the doctor’s office. Instead, she heard someone who sounded like her daughter crying as a man shouted in the background: “Lay down and put your head back.”
“Mom, these bad men have me. Help me! Help me!” she thought she heard her daughter plead. DeStefano bargained with the caller, who demanded US$1 million, but eventually settled on US$50,000.
Before the exchange of money, another parent warned DeStefano that this could be an AI scam. Eventually, DeStefano was able to reach her daughter, who confirmed she was safe. The entire ordeal, which DeStefano testified about before a US Senate judiciary committee, turned out to be an AI-generated hoax. DeStefano believes her daughter’s voice had been cloned using synthetic audio technology.
The American mother’s nightmare is now becoming a reality in Hong Kong. As one of the world’s most digitally connected cities, Hong Kong faces similar threats from increasingly powerful AI deception. Our city is no stranger to online scams. According to official statistics, deception caused residents to lose HK$8.52 billion (US$1.1 billion) in the first 11 months of 2024 alone.
AI has created a dangerous new frontier, supercharging online scams by enabling the cloning of voices and faces with a deceptive degree of realism. This sharply lowers the cost of online deception, enabling fraud at rapid scale. The technology behind deepfakes has advanced so quickly that we lack effective tools to separate falsities from facts.
A few months ago, Hong Kong radio and television host Sammy Leung Chi-kin received a video call from a person pretending to be actress Sandra Ng Kwan-yue, who claimed that “a director had sold her to Cambodia” and asked Leung for HK$5,000 so she could “buy a ticket for a boat ride back to Hong Kong”. Though Leung was discerning enough to see through the AI ruse, many others, especially the elderly and young, might lack the technological proficiency and awareness to recognise such falsehoods.
Visited 1 times, 1 visit(s) today
