AI Voice Scams: How Deepfake Calls Work and How to Protect Yourself

Your phone rings. It sounds exactly like your daughter. She is panicking, saying she has been in a car accident and needs money right now. Your heart races. You are about to send the money — but it is not actually her. It is an AI-generated clone of her voice.

This scenario is no longer science fiction. AI voice cloning technology has advanced to the point where a convincing replica of someone's voice can be generated from just a few seconds of audio pulled from social media videos, voicemails, or public recordings. Scammers are using this technology to exploit the one thing people trust without question: a familiar voice.

How AI Voice Cloning Works

Modern AI voice cloning requires surprisingly little source material. A short video posted on Instagram, a voicemail greeting, or even a brief phone call can provide enough data.

The process:

1. Audio collection — The scammer finds audio of the target's voice on social media, YouTube, TikTok, or through a brief phone call
2. Voice model training — AI software analyzes the voice's pitch, tone, cadence, and speech patterns
3. Real-time synthesis — The scammer can now generate speech in the cloned voice, sometimes in real time during a phone call

The technology that enables this is commercially available. Some tools require as little as three seconds of sample audio to produce a usable clone.

Common AI Voice Scam Scenarios

The Emergency Call

The scammer calls a parent or grandparent, impersonating a family member in distress. The "family member" claims to have been in an accident, arrested, or kidnapped and urgently needs money wired immediately.

The Boss Impersonation

An employee receives a call that sounds exactly like their CEO or manager, instructing them to make an urgent wire transfer or purchase gift cards for a client.

The Kidnapping Hoax

The caller claims to have kidnapped a loved one and puts a cloned voice on the phone as "proof." The fake victim cries for help while the scammer demands ransom.

Red Flags to Watch For

Even the best AI clones are not perfect. Watch for these warning signs:

Red Flag	Why It Matters
Extreme urgency	Scammers pressure you to act before you think
Request for unusual payment methods	Wire transfers, gift cards, and cryptocurrency are untraceable
"Don't tell anyone"	Isolation prevents you from verifying the story
Background noise or audio glitches	AI-generated audio may have unnatural pauses or artifacts
Emotional manipulation	Fear and panic override rational thinking
Caller avoids specific questions	The AI cannot answer detailed personal questions accurately

How to Protect Yourself and Your Family

1. Establish a Family Code Word

Agree on a secret code word or phrase that only your family members know. If anyone calls claiming to be a family member in an emergency, ask for the code word. A scammer using a cloned voice will not know it.

Tips for choosing a code word:

• Use something unrelated to personal information available online
• Avoid names of pets, birthdays, or addresses
• Change it periodically
• Make sure every family member memorizes it

Store the code word securely. A password-protected memo on LOCK.PUB is a good option — share the link with family members and set it so only those with the password can access it.

2. Always Verify Independently

If you receive a distressing call from someone who sounds like a family member:

1. Hang up. Do not stay on the line.
2. Call the person directly using the number saved in your contacts — not a number provided by the caller.
3. Contact another family member to confirm the person's whereabouts.
4. Wait. Scammers create urgency specifically to prevent you from verifying.

3. Limit Voice Exposure on Social Media

The less audio of your voice available publicly, the harder it is for scammers to clone it.

• Set social media accounts to private
• Avoid posting long videos with clear voice audio
• Be cautious about voice messages in public groups or forums
• Review privacy settings on platforms where you post video content

4. Be Skeptical of Unexpected Calls

Legitimate emergencies rarely require immediate wire transfers. No hospital, police station, or embassy will ask you to send money via gift cards or cryptocurrency.

5. Educate Vulnerable Family Members

Older relatives are frequent targets. Have a direct conversation with parents and grandparents about AI voice scams. Make sure they know:

• AI can now mimic anyone's voice convincingly
• They should always verify through a second channel before sending money
• It is okay to hang up and call back on a known number

What to Do If You Suspect a Voice Scam

1. Do not send money. No matter how convincing the voice sounds.
2. Hang up and verify by calling the real person directly.
3. Report the incident to local law enforcement and the FTC (in the US) or equivalent authority.
4. Alert your family so they can be on guard for similar attempts.
5. Document everything — save call logs, phone numbers, and any recordings if possible.

Protecting Your Family's Private Information

Beyond code words, consider how your family shares sensitive information in general. Passwords, account numbers, and personal details sent through regular messaging apps like iMessage or Messenger can be accessed if a device is compromised.

For sharing sensitive family information — emergency contacts, account details, medical information — use a password-protected link with an expiration date. LOCK.PUB lets you create encrypted memos that self-destruct, keeping private family data out of permanent chat histories.

The Bigger Picture

AI voice cloning is just one tool in a rapidly expanding arsenal of AI-powered scams. As the technology improves, the line between real and fake will become harder to detect by ear alone. The best defense is not better detection — it is better verification habits.

Make verification the default, not the exception. Establish code words, verify through known channels, and never let urgency override caution.

Create a Secure Family Code Word Memo →