Morgan Wright Warns $5,000 AI Voice-Cloning Scam Shows Skepticism Beats Software

Q: If scammers can clone your child's voice in seconds, what secret strategy can your family use to expose the fraud instantly?

Use a family code word or short secret phrase that only close relatives know. Experts cited in the reporting say this is one of the fastest, most reliable ways to defeat an AI voice-cloning scam, because a scammer may copy a voice but will not know a private family password or inside-joke phrase. The rule works only if the family agrees in advance to use it for any unusual emergency call, especially one involving panic, secrecy, or a demand for money. If someone claiming to be your child calls crying for help, ask for the code word immediately. If they cannot provide it, treat the call as a likely fraud, hang up, and contact your child through a trusted number or another family member. That matters because these scams are designed to override judgment. Morgan Wright said criminals use AI to create urgency so victims act before thinking. In the California case involving Deborah Del Mastro, scammers allegedly used an AI clone of her daughter’s voice in a fake kidnapping scheme and extracted more than $5,000. A code word should be easy for relatives to remember but hard for outsiders to guess. It should not be something visible on social media, such as a pet’s name, birthday, school, or favorite team. Families should also rehearse the response: pause, demand the code word, end the call if anything feels wrong, and verify through a separate channel. That combination of a secret phrase and calm verification is the strongest practical defense.

1 articles · Updated · Fox News · Oct 7

$5,000 was lost by a California mother after scammers cloned her daughter's voice in a fake kidnapping call, a case Morgan Wright cited to show how convincing AI fraud has become.
Wright said the strongest defense is to pause and verify, because scammers use urgency, fear and authority cues—not just technology—to push victims into acting before thinking.
A few seconds of audio scraped from social media can now power realistic voice clones, while generative AI also removes language barriers and drafts persuasive phishing messages at scale.
Detection tools are struggling to keep up, Wright said, because models from OpenAI, xAI, Anthropic and Gemini change constantly, making software-based defenses a moving target.