Updated
Updated · BBC.com · Aug 5
Families Adopt Secret Codewords Against $6.3 Billion AI Deepfake Scams
Updated
Updated · BBC.com · Aug 5

Families Adopt Secret Codewords Against $6.3 Billion AI Deepfake Scams

3 articles · Updated · BBC.com · Aug 5

Summary

  • $6.3 billion in AI-enabled scam losses in 2025 has pushed experts and victims to urge families to set secret codewords for emergency calls, a low-tech check that deepfake voice fraudsters cannot easily fake.
  • Elizabeth Benz nearly drove to pay $7,500 after a caller cloned her 16-year-old son's voice and threatened to kill him, until relatives reached the real boy at a football game.
  • Experts say the codeword works only if families rehearse using it under stress, because panic can stop victims from asking for it even when a password already exists.
  • Three warning signs recur in these scams: extreme urgency, demands for untraceable payment such as cash, gift cards or Bitcoin, and pressure to avoid contacting anyone else.
  • The advice comes as Americans lost an estimated $148.2 billion to online scams and cybercrime in 2025, up nearly 26% from a year earlier.

Insights

How did AI voice scams evolve from targeting terrified parents to draining millions from multinational corporations in just a few years?
Why did a prepared mother's family password fail against an AI voice clone, and how can you avoid the same mistake?