She heard her daughter crying. It wasn't her daughter.

A woman in her late 60s picks up the phone on a Tuesday morning. The voice on the other end is crying. It sounds exactly like her daughter. "Mom, I was in an accident. I'm okay but I need you to send $2,000 to this link right now. Please don't call Dad. Please don't call anyone. Just do this for me."

She sends the money.

Her daughter calls her that afternoon about dinner plans. She had no idea anyone had used her voice.

This is not a story from a science fiction screenplay. This is the most common version of a scam that the Federal Bureau of Investigation (FBI) formally identified in its 2025 Internet Crime Report as one of the fastest-growing fraud categories in America. The FBI logged 22,364 complaints tied to AI-assisted fraud in 2025. Reported losses totaled $893 million. And those are only the people who reported it.

This can literally happen to a loved one.

What AI Voice Cloning Actually Is

AI voice cloning is exactly what it sounds like. A criminal takes a short audio sample of someone's real voice, feeds it into a freely available software tool, and produces a synthetic version of that voice that can say anything the criminal types.

The audio sample does not have to be long. Research from McAfee Labs found that a free tool needed only three to four seconds of audio to produce an 85 percent voice match. Three seconds. That is one sentence from a TikTok video. That is the first word of a voicemail greeting. That is a clip from an Instagram Reel your teenager posted last month.

Once the clone is built, the criminal calls a parent or grandparent, plays the synthetic voice describing a crisis, and demands immediate action. The call often includes a second voice playing an authority figure: a police officer, an attorney, a bail bondsman. The target is instructed not to call anyone else. The money must move right now before the window closes.

The Federal Trade Commission (FTC) issued a formal consumer alert about this specific scam in March 2023. In 2025, the FBI documented it as a distinct category in its annual crime report for the first time. The losses that year from AI voice-clone distress calls alone exceeded $5 million.

Why This Scam Works So Well

The answer is not that victims are gullible. The answer is that the scam is designed to defeat the exact defenses we rely on most.

Most people have a threshold question when something feels off: does this sound like the person I know? With a cloned voice, the answer is yes. Every pitch, every cadence, every verbal habit that belongs to your child or grandchild is present in the call. The rational brain, which might otherwise pause and ask questions, receives a confirming signal from the most trusted source it has: the sound of a familiar voice under stress.

The FTC has found that people 60 and older bear the largest financial burden from fraud of all kinds. In 2025, over 50 reported $4.3 billion in total fraud losses. Older adults reported over $352 million in AI-related fraud losses alone, per the FBI. These are not numbers that describe careless people. They describe people who heard their grandchild's voice and responded the way any loving family member would.

The scam also exploits urgency. A criminal who gives you time to think gives you time to verify. The instruction to act immediately and tell no one is not incidental. It is the architecture of the attack.

Where Scammers Get the Audio

This is the part most people do not expect. The criminal does not need to hack your family member's phone or intercept a call. They need publicly available audio. That means:

1. TikTok videos, Instagram Reels, YouTube clips, and Facebook Stories posted by your family member

2. Voicemail greetings on a personal or business phone

3. Podcast appearances, live streams, or recorded interviews

4. Voice notes sent in public or semi-public group chats

5. Any video where the person's voice is clearly audible for three or more seconds

The FBI recommends setting social media accounts to private as a baseline precaution. It does not eliminate the risk entirely, but it removes the easiest harvesting targets. A criminal looking for audio samples will move to the next family if yours is not publicly accessible.

The One Defense That Actually Works

Technology companies are working on detection tools. Regulatory agencies are tightening rules. Phone carriers are exploring voice authentication systems. None of these will protect your family in the next 30 days.

The defense that works right now costs nothing and takes ten minutes to implement. It is a family code word.

A family code word is a private word or short phrase that every member of your household knows and agrees to use in any emergency call. The rule is simple: if a family member calls in a crisis, the first thing you do is ask for the code word. A scammer using a cloned voice cannot answer it.

The code word only works if every family member knows it and understands the protocol. That means having the conversation now, before any call comes. It also means agreeing as a family that hanging up and calling back on a known number is never rude. It is required.

Five Steps to Set Up Your Family's Defense Today

1. Choose one private code word or short phrase. Pick something memorable but not obvious. Avoid birthdays, pet names, or anything a stranger could guess from your social media.

2. Tell every family member who might receive or make an emergency call. That includes grandparents, adult children, and any caregiver with access to family finances.

3. Set the rule in writing. Write it down and share it in a private family group chat or email thread. "If you call in a crisis, we ask for the code word. If you cannot give it, we hang up and call you back on your regular number."

4. Set all family social media accounts to private. Go through TikTok, Instagram, Facebook, and YouTube and switch public accounts to friends-only or private. This reduces the audio that scammers can harvest.

5. Report any attempt immediately. If you receive one of these calls, whether you sent money or not, report it to the FTC at ReportFraud.ftc.gov and to the FBI at ic3.gov. Every report helps law enforcement identify patterns and disrupt operations.

If Money Was Already Sent

Contact your bank immediately. Wire transfers and cryptocurrency transactions are difficult to recover, but some can be stopped if caught within hours. Gift card payments are almost always unrecoverable. Your bank's fraud team will tell you what options exist.

File a report with the FTC, the FBI, and your local police. When you file with the FBI at ic3.gov, note specifically that AI voice cloning was involved. The FBI's AI fraud statistics only count cases where victims flag it directly. Every report improves the accuracy of the data that law enforcement and regulators use to respond to this threat.

Do not carry shame about this. The person who fell for a cloned voice did not make a bad decision. They made the decision any caring family member would make when they believed someone they loved was in danger. Failure belongs to technology being weaponized against ordinary people, not to the person who responded to what sounded like a real emergency.

The Bigger Picture

AI voice cloning fraud is one visible example of how artificial intelligence is being used against people who have not been given the tools to recognize it. Technology is not going away. The scams are going to get more convincing, not less.

The goal of AI Educational Solutions is not to make every business owner into a cybersecurity expert. It is to give you enough practical knowledge to protect yourself, your family, and your business before a scam reaches you.

The code word is not a high-tech solution. It is a human-first one. And right now, it is the most effective defense available.

If you want to understand more about how AI is being used for and against businesses in 2026, the August 11 workshop is where that conversation is happening.

Has this scam happened to someone you know?

I didn’t write this article to scare anyone just bring attention to the scam. I wrote it because it happened to an individual I know. There will always be bad actors to use technology to defraud people. We need to remain vigilant. Do me a favor and set up a safe word today with the ones you love.

Family Code Word Protocol Generator

What this prompt does:

This prompt acts as a family safety advisor and walks the user through building a private, memorable code word system their family can use to verify any emergency call before anyone sends money or takes action. It produces a written protocol the family can share and refer back to.

Paste this prompt into ChatGPT, Claude, or any AI assistant:

You are a family safety advisor specializing in fraud prevention, specifically AI voice cloning scams. Your job is to help me build a Family Emergency Code Word Protocol that every member of my household will know and use.  Walk me through the following steps one at a time. Ask me each question, wait for my answer, then move to the next step. Do not rush through them.  Step 1: Ask me how many people are in my immediate family or household who would be part of this protocol.  Step 2: Ask me to list the names or relationships of the family members who need to know the code word (examples: Mom, Dad, Grandma, teenage son).  Step 3: Help me choose a code word or short phrase. Suggest three options based on criteria that make a good code word: memorable, not obvious to outsiders, not tied to publicly known details like birthdays or pet names, and easy to say clearly under stress. Ask me to choose one or suggest my own.  Step 4: Write out the Family Emergency Code Word Protocol as a short, clear document I can share with my family. Include: the code word I chose (marked as a blank I fill in before sharing), the rule for when to use it, what to do if the caller cannot provide it, how to verify a real emergency after hanging up, and where to report a scam attempt.  Step 5: Draft a short message I can send to each family member to introduce the protocol. Keep it calm and clear. Do not make it scary. Make it feel like a practical household rule, not a crisis.  Step 6: Suggest two ways my family can practice the protocol before a real call comes, so that using it feels natural rather than awkward.  After completing all six steps, give me a one-paragraph summary of what we built and why it works.

Why this prompt matters:

The FBI's 2025 Internet Crime Report documented $893 million in AI-related fraud losses. Grandparent and family emergency scams, powered by cloned voices, accounted for over $5 million in reported losses in that same year. The average loss per victim was $11,000. A family code word costs nothing and takes ten minutes to implement. This prompt makes sure the conversation actually happens.

Michael Carmine
Founder & CEO | AIEducationalSolutions.org
About the Author

Michael Carmine, Ed.D., MBA is the Founder and Chief Executive Officer of AI Educational Solutions (AIES), a boutique AI consulting and training firm based in Chapin, South Carolina, serving small and medium businesses across the Midlands and nationally. He holds a doctorate in Curriculum and Instruction with a specialty in Instructional Design and Technology, and has trained over seven hundred professionals across the United States, South Korea, and Qatar. Most AI consultants come from technology. Michael comes from education, and that difference shapes every tool he builds.