"Crying Child" AI Kidnapping Scams and the New Wave of Family Emergency Fraud

ai-securityfraud-detectionmitre-f3mitre-attck

What’s Happening

AI-powered family emergency scams have escalated dramatically in June 2026. A deeply disturbing new variant — the “crying child” scam — uses AI voice cloning to make parents believe their child has been kidnapped, injured, or arrested. F-Secure, the cybersecurity firm, reports that 56% of people now encounter online scams monthly, and 52% of victims lose money to them PCMag.

The numbers are staggering. The FBI’s 2025 Internet Crime Report recorded $893 million in AI-related scam losses across 22,364 complaints Malwarebytes. Deepfakes now account for 11% of all global fraudulent activity, according to Sumsub’s 2025-2026 Identity Fraud Report Sumsub. Google’s Trust & Safety team issued a June 2026 advisory flagging AI cryptocurrency investments, Adversary-in-the-Middle (AITM) attacks, and mobile scams as the top threats right now Google Blog.

How It Works — The Five Newest Attack Vectors

1. The “Crying Child” AI Kidnapping Scam

This is this month’s most alarming development. Scammers extract as little as 10 seconds of audio from your child’s social media videos — TikTok, YouTube, Instagram Stories, school performances posted online. They run it through AI voice cloning tools to generate a terrified voice saying they’ve been kidnapped or injured.

The victim’s phone rings. They hear what sounds like their child crying, screaming, or calling for help. Then a “kidnapper” or “police officer” gets on the line demanding cryptocurrency or wire transfers for ransom. Cybersecurity researcher Megan Squire at F-Secure warns that most AI voice cloning tools need only 10 seconds of clear audio to produce a convincing clone PCMag.

Bitdefender documented this exact scam on June 24, reporting that scammers pull audio from public social media videos, school performances, sports clips, and YouTube content — then use AI to make it say anything they want Bitdefender.

2. AI-Generated Fake Shopping Sites

The Guardian reported on June 7 that scammers are using ChatGPT to create entire fake storefronts — cloned versions of legitimate retailers with slightly altered domain names. These AI-generated sites look professional, have realistic product catalogs, and accept credit cards. The customer receives either nothing or a counterfeit item The Guardian.

Worse, some of these fake stores show up in AI chatbot responses. When users ask ChatGPT for product recommendations, it can surface these scam sites alongside legitimate ones.

3. Adversary-in-the-Middle (AITM) Phishing

Google’s June 2026 advisory warns about AITM phishing kits that intercept two-factor authentication codes in real time. The scammer sets up a fake login page that forwards credentials to the real site — but captures the 2FA code as the user types it. This defeats SMS-based two-factor authentication entirely Google Blog.

4. AI Cryptocurrency Investment Scams

Google flags AI-themed crypto investments as one of the fastest-growing scam categories. Scammers use AI-generated videos of celebrities and “financial experts” endorsing fake trading platforms. The platforms show impressive AI-generated performance dashboards and allow small withdrawals — building trust before the “rug pull” where all deposited funds vanish Google Blog.

Chainalysis reports $14 billion in crypto scam losses in 2025, with AI-enabled scams proving 4.5 times more profitable than traditional fraud Vectra AI.

5. The “Hi Mum” Text Scam — Now AI-Powered

A variant that started in the UK and is now global: the victim receives a text from an unknown number saying “Hi Mum, I lost my phone, this is my new number.” The scammer impersonates the child via text, then eventually asks for money to pay an urgent bill. With AI, these texts are now personalized — they know the child’s name, school, and other details scraped from social media. F-Secure researchers note that older parents with adult children are targeted just as frequently as parents of young kids PCMag.

Real Examples

The $893 million FBI tally: The 2025 FBI Internet Crime Report logged 22,364 AI-related complaints, with total losses of $893,346,472. The agency warns that voice cloning and deepfake scams grew faster than any other category Malwarebytes.

The grandfather who lost $6,000: F-Secure’s PR representative shared that her own grandfather received a call from someone claiming to be a lawyer, saying his grandson had been in a motorcycle accident and needed $6,000 for legal fees. The grandfather sent the money — and only called for a second opinion afterward PCMag.

The Florida mother conned out of $15,000: In July 2025, a Florida mother received a call that appeared to be from her daughter, claiming she had been arrested after a car accident. Someone claiming to be a lawyer demanded $15,000 for bail. The mother paid in cash before a request for further funds tipped off a relative Sumsub.

The California mom AI voice-kidnapping: In May 2026, a Bay Area mother received a call with her daughter’s cloned voice saying she had been kidnapped. The scammers demanded thousands of dollars. The case was covered by ABC7 and highlights how realistic AI voice clones have become ABC7.

Sophisticated fraud up 180%: Sumsub’s report shows a 180% increase in “sophisticated fraud” — AI-generated identities, deepfake-powered social engineering, and autonomous fraud agents — compared to 2024 Sumsub.

Red Flags — What to Watch For

Red Flag What Scammers Want You to Do
Call from “your child” crying for help Send cryptocurrency ransom immediately
“Lawyer” or “police officer” calls about a family emergency Wire money for bail or legal fees
Text from “Mum/Dad, I lost my phone” with a new number Send money for an “urgent bill”
AI chatbot recommends a shopping site you’ve never heard of Enter credit card details on a fake store
“Urgent: your account is compromised” with a login link Enter credentials + 2FA code on a fake page
Celebrity-endorsed crypto investment ad on social media Deposit funds into a fake trading platform

Critical rule: AI-generated voices are now indistinguishable from real ones to the human ear. Academic research cited by Sumsub found that participants correctly identified real vs. AI voices only 37.5% of the time Sumsub. You cannot trust what you hear. You must trust your verification process.

How to Protect Yourself and Your Family

Create a family code word — and never type it anywhere

Pick a passphrase that only immediate family members know. If someone calls claiming to be a relative in distress, ask for the code word. If they can’t answer, it’s a scam. The code word must never be written in emails, texts, or social media messages — scammers data-mine those.

If you receive a family emergency call, do this immediately

The three-step response, recommended by F-Secure researcher Megan Squire PCMag:

  1. Stop — Don’t react. The urgency is the scam’s weapon. Take a breath.
  2. Call your child directly — Hang up and call their real number. In almost every case, they answer, confused and unharmed.
  3. Verify through a separate channel — If they don’t answer, call another family member, the school, or their workplace. Never use the number the scammer gave you.

Lock down your family’s social media

Scammers mine public social media for voice samples and personal details. Make your profiles private. Review what videos of your children are publicly accessible. Even a 10-second clip from a school concert is enough to clone a voice. F-Secure notes that scammers don’t need to know you personally — a quick online search reveals names, relationships, and enough detail for a convincing story PCMag.

Use an authenticator app, not SMS 2FA

Google’s advisory specifically warns that SMS-based two-factor authentication can be bypassed by AITM phishing kits. Use an authenticator app (Google Authenticator, Microsoft Authenticator, Authy) or a hardware security key for your most important accounts — email, banking, crypto exchanges Google Blog.

If ChatGPT or another AI chatbot recommends a store you’ve never heard of, manually type the URL into your browser and check for reviews on independent sites. Look for signs of legitimacy: a physical address, customer service phone number, and presence on well-known review platforms. The Guardian reports these fake storefronts are multiplying rapidly The Guardian.

Freeze your credit

Freezing your credit with the three major bureaus (Equifax, Experian, TransUnion) is free and takes 15 minutes each. It prevents fraudsters from opening accounts in your name even if they have your Social Security number.

Run background checks on investment platforms

Before depositing money into any AI-themed investment platform, search for it with “scam” or “complaint.” Check your state securities regulator. Verify the platform is registered with the SEC. AI-generated investment dashboards and celebrity endorsements are trivial to fake.

What to Do If You or a Family Member Has Been Targeted

  1. Stop all communication with the scammer. Do not argue, threaten, or engage — just disconnect.

  2. Report it immediately:

    • FBI IC3 at ic3.gov — for any AI-related scam
    • FCC at fcc.gov/complaints — for voice cloning and phone scams
    • SEC at sec.gov/complaint — for investment fraud
    • Local police — especially for kidnapping/extortion threats

  3. Contact your bank right now. Wire transfers and cryptocurrency payments are often irreversible, but banks can sometimes flag accounts and freeze funds if you act within hours.

  4. Change all passwords on compromised accounts. Use a password manager to generate unique, strong passwords for every account. Enable authenticator-app 2FA on every service that supports it.

  5. Monitor your credit at annualcreditreport.com (free weekly reports from all three bureaus). Set up a fraud alert by contacting any one bureau — they notify the other two automatically.

  6. Talk about it. AI family-emergency scams thrive on shame and silence. Many victims don’t report because they feel foolish. The more we discuss these tactics openly — at dinner tables, in group chats, in community forums — the harder it is for scammers to find new victims. F-Secure’s Squire puts it plainly: “You’ll feel bad if you question your child and they’re really fine. You’ll feel worse if you just lost $30,000” PCMag.

Sources

  • FBI 2025 Internet Crime Report via Malwarebytes (June 8, 2026) — malwarebytes.com
  • PCMag: “This Nightmare AI Scam Is Making Parents Believe Their Child Has Been Abducted” (June 5, 2026) — pcmag.com
  • Bitdefender: “The ‘Crying Child’ Scam: How Scammers Use AI Voice Cloning to Target Parents” (June 24, 2026) — bitdefender.com
  • Google Blog: “Our Latest Fraud and Scams Advisory” (June 8, 2026) — blog.google
  • The Guardian: “Cloned sites: the shopping scams that lead ChatGPT to fake stores” (June 7, 2026) — theguardian.com
  • Sumsub: “Fraud Trends 2026: AI Scams, Deepfakes, and Emerging Threats” (April 15, 2026) — sumsub.com
  • Vectra AI: “AI Scams in 2026: How They Work and How to Detect Them” — vectra.ai
  • ABC7 News: “Mom out thousands after scammers use AI to mimic daughter’s voice” (May 26, 2026) — abc7ny.com
  • CNN: “AI ‘voice cloning’ scams are on the rise” (May 29, 2026) — cnn.com
  • Forbes: “AI Scams Are Rising In 2026 — Here’s How to Protect Yourself” (June 15, 2026) — forbes.com

📖 Related Reads

Cross-links automatically generated from None.