Deepfake ‘Kidnapping’ Calls and the Safe Word Fix

We Live Security · High sophistication
Last updated September 10, 2026

The article describes how scammers use AI voice cloning to impersonate a loved one during a phone call and pressure relatives into paying money (often framed as a kidnapping emergency). It recommends a practical defense: a pre-agreed family “safe word” and a simple verification process (hang up and call back on a known number) to break the scam’s urgency.

Key findings

  • The article claims deepfake voice calls are increasingly common and hard for many people to detect.
  • Scammers can clone a voice with only a few seconds of audio, often scraped from social media or online work content.
  • A prominent workflow described is a ‘virtual kidnapping’ call using a cloned voice briefly, with added sobbing/background noise and personal details to increase believability.
  • A pre-agreed safe word and a call-back-to-known-number procedure are presented as simple, effective controls to stop the scam.
  • Victims are warned to watch for “secondary fraudsters” offering paid recovery services after the initial scam.

Who’s being targeted

  • Commonly targeted roles: All employees, Executives, Finance (for payment/wire skepticism), HR and Communications (staff with public-facing content).
  • Affected industries: Consumers/Households, General public (individuals and families).
  • Attack channels: vishing.
  • Impersonated: A family member (voice-cloned deepfake).

Awareness takeaways

  • Assume a convincing voice on the phone could be fake; verify with a safe word or a call-back to a known number.
  • Reduce voice-cloning risk by limiting public access to videos/audio and tightening social media privacy settings.
  • Watch for emotional manipulation tactics (urgency, distress sounds, chaos) designed to stop you from thinking clearly.
  • After a scam, stop contact, preserve evidence, notify your bank, and be alert for ‘recovery’ scams.

Red flags to watch for

  • Caller creates extreme urgency and emotional distress to stop verification
  • Voice is only used briefly and masked with ‘sobbing noises’/background noise
  • Pressure to act without calling back via a known number
  • Caller refuses, dodges, or becomes angry when asked for the safe word
  • Caller insists you must stay on the line and not call back
  • Caller pushes for money or sensitive info instead of verification
Try Mirage

Mirage safely runs attacks like this one against your own team, so you find out what happens before a real adversary does.

Get a demo
Read the video transcript

Your phone rings. You hear your daughter’s voice: “Hi… please help… I’ve been kidnapped…” then sobbing and chaos. Here’s the twist: that voice can be a deepfake, cloned from a few seconds of audio scraped off social media, used for just moments, buried under crying and noise to make you panic and pay fast. The giveaway isn’t the sound, it’s the behavior: extreme urgency, pressure to stay on the line, and refusal to let you hang up or call back on a known number from your contacts. Fix it now: agree on a family safe word. If a call feels wrong, ask for it once, then hang up and call back using the number saved in your phone, no matter how real the voice sounds.

Similar attacks

Deepfake Video Call Drove $25M Wire Transfer Scam

Deepfake Video Call Drove $25M Wire Transfer Scam

The article discusses Google’s new selfie-video account recovery, but it also highlights a real deepfake-enabled fraud case. In that incident, a finance employee joined a video call showing deepfake versions of coworkers and was persuaded to send multiple wire transfers, illustrating how realistic…

July 23, 2026
Deepfake FBI Videos Push Victims to Fake IC3 Sites

Deepfake FBI Videos Push Victims to Fake IC3 Sites

The FBI warned that scammers are impersonating IC3 leadership using AI-generated (deepfake) videos and spoofed IC3 websites to trick prior fraud victims into sharing more personal and financial information. In one example, victims are contacted on Facebook Messenger by someone posing as an FBI…

July 21, 2026
AI Vishing Works Because Scripts Persuade

AI Vishing Works Because Scripts Persuade

Researchers tested AI and human voice scam calls and found people comply mainly because the caller’s script is persuasive, not because the voice sounds perfectly human. Even when listeners correctly suspect the voice is synthetic, many still continue the conversation and may hand over sensitive…

July 17, 2026
Fake Google & Gemini Calls Led to $240M Bitcoin Heist

Fake Google & Gemini Calls Led to $240M Bitcoin Heist

A group of young scammers stole more than $240 million in bitcoin by calling a wealthy crypto investor and impersonating trusted companies. The callers claimed the victim’s accounts and wallet were under attack, then tricked him into granting access and sharing security codes that enabled the…

September 8, 2026
Fake IT Helpdesk Tricks Users Into Remote Access

Fake IT Helpdesk Tricks Users Into Remote Access

This bulletin describes multiple real-world social engineering campaigns where attackers impersonate IT support or use trusted-looking sharing and “Allow” prompts to gain access. Several campaigns abuse Microsoft Teams and document-sharing lures to trick employees into installing remote tools or…

September 3, 2026
Call-Blocker Seller Fined for Cold-Calling Elderly

Call-Blocker Seller Fined for Cold-Calling Elderly

UK regulator ICO fined Elderly Aids Ltd £190,000 for making over 758,000 unsolicited marketing calls to people who had registered to avoid such calls. Complainants said the callers used aggressive, misleading tactics and sometimes failed to identify themselves, pressuring elderly targets into paid…

August 27, 2026