AI Voice Cloning Scams: How to Spot and Stop Them

An AI voice cloning scam is a fraud in which a criminal uses software to imitate a real person’s voice, often a family member, boss, or official, to pressure someone into sending money or sharing sensitive information. Because cloned voices can sound convincing, the safest defense is to verify through a separate, trusted channel before you act.

Most scam calls still do not use cloned voices, and the same habits protect you either way. Caller ID can be faked, so a familiar name or number proves nothing. Detecting a fake by ear or by app is unreliable, so verifying and slowing down work better. Payment method matters, because some payments are very hard or impossible to recover.

Think you’re being scammed right now? Hang up. Don’t send money, codes, or gift cards. Call the person on a number you already have saved. If money has moved, contact your bank or payment provider immediately. Report at ReportFraud.ftc.gov and, for internet-enabled crimes, ic3.gov. Steps are in the recovery section.

Disclosure. This is educational content, not legal, financial, or law-enforcement advice. Whether you get money back depends on the payment method, the facts, and the institution’s policy. No technique reliably detects every cloned voice.

Key takeaways

• Verify through a channel the caller did not give you. Hang up and call back on a number you already have.

• Urgency, secrecy, and unusual payment requests are the main warning signs, whether or not a voice is cloned.

• Caller ID, familiar voices, and “AI detector” results are not proof of anything.

• A family safe word helps, but it works best alongside a callback.

• Gift cards, wires, crypto, and cash handoffs are the hardest payments to recover.

• If money moved, contact your payment provider first, then report to the FTC and, where relevant, the FBI’s IC3.

• Being targeted is not a personal failing. Scammers exploit trust and time pressure.

Methodology and disclosure. This guide is current as of November 2026. Its research was done in November 2026. It draws on government alerts, official data, and published research, and prioritizes sources in this order: government and regulators, laws and program rules, academic research, then established journalism. It did not test any product, and no tool, app, bank, or vendor is ranked or recommended. It contains no affiliate links. [VERIFY: publisher to confirm compensation and conflict disclosure before publication.] Scam tactics, laws, and statistics change, so confirm important details with the official sources listed at the end.

1. What Is an AI Voice Cloning Scam?

Short answer: It is an impersonation scam that adds software-made speech to a familiar fraud script. The goal is the same as in older scams: money or information, delivered fast.

Voice cloning uses software to make speech that resembles a real person. In 2023, the Federal Trade Commission (FTC) warned that a scammer may need only a short audio clip of a relative, possibly taken from online content, plus a voice-cloning program. Voice cloning is one piece of a larger toolkit, and it is easy to confuse with related terms. knowbe4

What this means for you. A cloned voice, a spoofed number, and a scripted story can all appear in one call, and each can be fooled independently. That is why this guide centers on verification, not on spotting the fake.

2. How Do AI Voice Cloning Scams Work?

Short answer: Scammers gather a voice sample, build a believable story, and create time pressure so you act before you check.

At a defender’s level, the pattern has three parts:

1. Source material. Audio can come from public videos, social posts, or voicemail greetings. The FTC’s 2023 alert described clips posted online as a source. The FBI has also noted that the more information is public about you, the more scammers can use it to make a story believable. Published estimates of how much audio is “enough” vary, and quality differs, so this guide does not give a number. knowbe4greenepublishing

2. A story. The caller claims a crisis (an accident, an arrest, a stranded trip) or a business emergency (an urgent payment).

3. Pressure. Expect secrecy requests, “don’t hang up,” and unusual payment methods.

Why it works. Scammers exploit trust and urgency, not carelessness. A frightened person hearing a loved one’s voice has every reason to act. Careful, tech-savvy people can be fooled too.

3. What Are the Common Types of Voice Cloning Scams?

Short answer: Family emergencies, workplace payment requests, fake officials, “virtual kidnapping,” and fake endorsements are the patterns that federal agencies have described.

Illustrative scenario 1: family emergency call. This scenario is hypothetical and uses no real people. A retiree gets a call from a voice that sounds like a grandchild. The caller says there was an accident, begs her not to tell anyone, and asks for payment by gift card today. The red flags are secrecy, urgency, and a payment method that is hard to reverse. A safe response: say “I’m going to hang up and call you back,” then call the grandchild’s saved number and, if there is no answer, a parent or sibling. A real emergency can survive a short pause.

Illustrative scenario 2: workplace payment request. Hypothetical. A finance employee gets a voicemail that sounds like a manager asking for an urgent wire to a new account. Company procedure should settle this: call the manager back on the number in the company directory, require a second approver, and allow no “urgent” exceptions. If the request is real, the check costs minutes. If it is not, it saves the payment.

Illustrative scenario 4: robocall in a public figure’s voice. Hypothetical. A voice message sounds like a well-known official asking you to “move this conversation” to another app. Don’t click or reply. Look up the person’s or agency’s official contact information yourself. The FBI has warned that, since at least 2023, malicious actors have sent text and AI-generated voice messages claiming to be senior U.S. officials. The warning was updated in December 2025. lilifepolitics

A note on “virtual kidnapping.” The FBI describes these as extortion scams in which no one has actually been taken. In December 2025 it warned that criminals may use altered photos from social media as fake “proof of life.” Common signs include calls that don’t come from the person’s phone, callers who work to keep you on the line, and ransom accepted only by wire. If this happens, try to reach the person directly or through someone near them, and contact authorities. newswavnbcchicago

4. How Common Are These Scams? What Does the Data Show?

Short answer: Reported losses to impersonation fraud are large and growing, but no official source cleanly separates “voice cloning” from other impersonation. All figures below are reported losses, not total losses.

• FTC (2025 data): Reported fraud losses reached a record $15.9 billion, up from $12.5 billion in 2024. Imposter scams drew more than 1 million reports, with losses of $3.5 billion. The FTC notes that reported losses are only a fraction of actual losses. Americans lost a record $15.9 billion to scams in 2025, FTC says +2

• FBI Internet Crime Complaint Center (IC3), 2025 Internet Crime Report (released April 2026): IC3 received 1,008,597 complaints and reported $20.9 billion in losses. For the first time the report broke out AI-related complaints: 22,364 complaints and $893,346,472 in losses. The FBI cautioned that actual figures may be higher. pymntsdmarcian

• Business email compromise (BEC): IC3 recorded about $3.05 billion in BEC losses in 2025. Coverage of the report says AI-linked BEC losses exceeded $30 million. That coverage describes voice cloning as used in follow-up calls that appear to come from an executive and reinforce written wire instructions. dmarciansecureworld

• Older adults: Adults 60 and over filed 201,266 IC3 complaints and reported $7.75 billion in losses in 2025, up 59%. These figures cover all internet crime, not only AI scams. housingwire

Limits. “AI-related” in IC3 data means a complaint referenced AI. It does not prove a voice was cloned. Reports undercount, and these numbers should never be read as the odds that any one call is a clone.

5. Can You Tell If a Voice Is Cloned?

Short answer: Not reliably. Plan as if you can’t, and verify instead.

In a 2023 peer-reviewed study (PLOS ONE) with 529 participants, listeners correctly spotted speech deepfakes only 73% of the time. Brief training improved results only slightly. The study used older synthesis methods, and participants knew they might hear a fake, which people on a real scam call would not. The same researchers found automated detectors unreliable when audio conditions change. Warning: Humans cannot reliably detect speech deepfakes +2

6. How Do You Verify a Suspicious Call?

Short answer: Hang up and contact the person through a channel you already trust. This works whether or not the voice is fake.

The FTC’s advice has been consistent: don’t trust the voice. Call the person on a number you know is theirs, and if you can’t reach them, try another family member or friend. knowbe4

Illustrative scenario 5: a verification script. Hypothetical. Calm phrases help because they end the call without an argument:

• “I’m going to hang up and call you back on the number I have.”

• “I’ll call your mom and call you right back.”

• “If this is real, a five-minute check won’t hurt.”

These work because they don’t accuse anyone, they move the conversation to a channel you control, and they don’t require you to judge the voice.

Family safe word. The FBI’s December 2024 guidance recommends creating a secret word or phrase with your family to verify identity. ic3

Illustrative scenario 6: safe-word setup. Hypothetical. Choose a word or phrase that isn’t posted online, isn’t a pet’s or street name, and is easy to remember. Share it in person or by a trusted channel, and agree that anyone asking for emergency money must give it. Be honest about limits: people forget it under stress, a scammer may claim they can’t remember it, and it works best alongside a callback.

7. How Do You Protect Yourself and Your Family?

Short answer: Build a few habits before you need them.

1. Slow down. Urgency, secrecy, and unusual payment methods are the main signals.

2. Hang up and verify. Use a saved number, not one the caller supplies.

3. Agree on a family safe word (see above).

4. Treat caller ID as untrustworthy. The FCC defines spoofing as deliberately falsifying the information shown on your caller ID. ctu

5. Never pay or share codes because of an inbound call. That includes one-time passcodes and passwords.

6. Lock down accounts. Use app-based or security-key two-factor authentication where offered, review bank alerts, and set transfer limits if your bank offers them.

7. Reduce, but don’t obsess over, public audio. Privacy settings help, but total removal isn’t realistic.

8. Prepare older relatives and teens. Use simple rules and no shame. Example: “No money on a phone call without a second person checking.”

9. Know where to report before you need to (see Section 12).

Parents and grandparents. Write the rules on paper: hang up, call back, tell someone, never gift cards. Practice once. Tell them plainly that a call from a loved one’s voice can still be a scam, and that checking is never rude.

8. How Do You Protect Your Business?

Short answer: Use written procedures that don’t depend on anyone recognizing a voice.

• Verify any payment request or bank-detail change by calling a number already on file, never one in the message.

• Require dual approval for wires above a set limit, and for all changes to payment details.

• Allow no “urgent” exceptions, including for executives.

• Train staff on voice and video impersonation, and let them know that refusing to bypass procedure is never penalized.

• Know your bank’s wire-recall contact before you need it.

FBI IC3 guidance on BEC has long advised confirming requests through a known number or channel and using two-factor controls, as summarized in press coverage of the FBI’s alerts. [VERIFY: link current IC3 BEC advisory.] krebsonsecurity

9. What About Bank Voice Authentication?

Short answer: Voice login alone is a risk. Where your bank offers other factors, use them.

Illustrative scenario 3: a voice-login scare. Hypothetical. A reader’s bank offers a voice-login option. They hear about cloning and wonder what to do. They can ask the bank what factors protect the account, enable app-based verification and transaction alerts if offered, and treat any “bank” call as unverified until they call back on the number on their card.

What official sources say: the FBI’s 2024 alert warned that AI audio could be used to obtain access to bank accounts. FinCEN’s November 2024 alert (FIN-2024-Alert004) pointed financial institutions to multifactor authentication, including phishing-resistant MFA, and live verification checks. That alert focuses on deepfaked identity documents, not on customer voice login. I did not find formal regulator guidance specific to voiceprint login. [VERIFY: current Federal Reserve, FFIEC, or CFPB statements on voice authentication.] For illustration, not an endorsement: at a July 22, 2025 Federal Reserve conference, OpenAI’s CEO said some institutions still accept voiceprints, and called that “crazy.” That is a tech executive’s opinion, not regulator guidance. ABA Banking Journal +2

10. Should You Use AI Detection or Call-Screening Tools?

Short answer: This article does not rank or recommend products, because detection tools are unreliable and change quickly. A verification-first approach needs no software.

The FTC’s Voice Cloning Challenge in 2024 found some promising ideas, but the agency said there is still no silver bullet. If you still want a tool, evaluate it with these questions. ftc

Over-trusting a detector is itself a risk. A “real” label can lull you; a “fake” label can wrongly accuse a real relative.

11. What Should You Do If You Already Sent Money or Shared Information?

Short answer: Act quickly, and start with whoever holds the money.

If it already happened, this is not your fault. Speed matters because some payments can’t be reversed once they clear. The FTC says to contact the company you used to send money and tell it the payment was fraud.

Adults 60 and older can also call the Justice Department’s National Elder Fraud Hotline, listed in local reporting as 833-FRAUD-11 (833-372-8311). [VERIFY: confirm on justice.gov.] loudounnow

Payment methods and recovery. Outcomes depend on the facts, the provider, and its policy. No official source guarantees reimbursement.

Illustrative scenario 7: time versus reversibility. Hypothetical, with no invented numbers. Ten minutes spent calling back and checking costs almost nothing if the call was real: you lose a few minutes. If the call was a scam, those ten minutes may be the difference between money still in your account and money that can’t be recovered. The cost of waiting is small, and the cost of acting too fast can be permanent.

12. What Are Your Rights, and Where Do You Report?

Short answer: Federal rules protect some losses and not others. Whether you were tricked or someone else took the money often decides which.

Keep four layers separate: (a) law and regulation, (b) regulator guidance, (c) payment-network and program rules, (d) individual firm policy.

• Regulation E (debit and electronic transfers). Reg E limits your liability for unauthorized electronic transfers, with a $50 limit if you report within two business days and $500 within 60 days, per consumer-finance summaries. [VERIFY with CFPB.] CFPB guidance in June 2021 said that if you’re tricked into sharing account access and a third party then initiates a transfer, it can count as unauthorized. But payments you yourself send after being deceived are generally not covered, even though some banks voluntarily reimburse. Whether a given scam fits is fact-specific and sometimes contested, and consumer advocates have called for changes. Don’t assume protection applies. ommbankrate

• Regulation Z (credit cards). Liability for unauthorized use is capped at $50, and billing errors must be reported within 60 days of the statement. The issuer must acknowledge within 30 days and resolve within two billing cycles, no more than 90 days. Whether a scam you consented to fits depends on the facts and the issuer’s and network’s rules.

• Reporting channels: the FTC at ReportFraud.ftc.gov; FBI IC3 at ic3.gov; the FCC consumer complaint center for spoofed or illegal calls; IdentityTheft.gov if personal data was exposed; your state attorney general; local police (which can help with bank paperwork). Reports help investigators and may not result in recovery. Be careful of “recovery” offers: IC3 says it never communicates directly by phone, email, or social media, and never asks for payment to recover funds. Readers outside the U.S. should use their own country’s national fraud-reporting agency. ic3

13. Is Voice Cloning Illegal?

Short answer: Cloning itself isn’t automatically illegal, but using it to defraud, impersonate, or place unconsented robocalls is.

The FBI says creating or sharing synthetic content is not inherently illegal, but it can facilitate crimes such as fraud and extortion. Here is where things stand as of November 2026: ic3

• FCC (federal). On February 8, 2024, the FCC ruled that AI-generated voices are “artificial” under the Telephone Consumer Protection Act, clarifying existing rules rather than creating new ones. Consent rules apply to such calls. It was not a total ban. The FCC later fined a consultant $6 million over AI-voiced robocalls spoofed to look like a prominent local figure, and a carrier agreed to a $1 million settlement. Those penalties rested on caller-ID rules and related duties. cooley

• FTC (federal). The Government and Business Impersonation Rule (16 CFR Part 461) took effect in April 2024. In February 2024 the FTC proposed extending it to impersonating individuals and to those who knowingly supply tools for it. As of this article, I found no final version of that extension. [VERIFY current docket status.] On September 24, 2026 the FTC announced a separate advance notice about platforms’ ad practices that may further impersonation scams. The FTC also says its Telemarketing Sales Rule covers AI-enabled scam calls. mayerbrown

• Caller-ID law. The Truth in Caller ID Act prohibits spoofing with intent to harm, defraud, or wrongfully obtain something of value. fcc

• State law varies. Tennessee’s ELVIS Act, effective July 1, 2024, extends its right-of-publicity law to voice, including simulated voices. Do not assume other states match; check your own. reedsmith

• Proposed federal law. The NO FAKES Act (S. 4591 / H.R. 8915), reintroduced May 20, 2026, would create a federal right over voice and likeness replicas. The Senate Judiciary Committee advanced it on June 18, 2026, but it is not law. letsdatascience

14. Common Mistakes

• Trusting caller ID.

• Trusting a familiar voice.

• Staying on the line to “figure it out.”

• Using the number the caller provides.

• Acting before telling anyone.

• Paying by gift card, crypto, or wire on a caller’s instruction.

• Sharing one-time codes.

• Assuming only older adults are targeted.

• Relying on “I can tell it’s fake” or on an AI detector.

• Skipping a family safe word, or never practicing it.

• Posting personal details that make a story easier to believe.

• Letting a business make “urgent” exceptions to wire procedures.

• Waiting to report.

• Blaming yourself and not reporting.

• Sharing a suspicious clip widely before checking it.

15. Common Misconceptions

16. Quick-Start Checklist

• Do we have a verification habit and a safe word?

• Are my parents’ and kids’ phones set up with scam protections?

• Does my bank offer transfer alerts and limits?

• Is 2FA on for email, bank, and phone accounts?

• Does my business have a callback and dual-approval rule?

• Do I know where to report?

• What will I say if the call comes?

This is a starting point, not a substitute for professional advice.

17. Glossary

• Account takeover: someone gains control of your account.

• Authorized vs. unauthorized transaction: whether you initiated a payment, which affects legal protections.

• Biometric: a body or voice trait used to identify you.

• Business email compromise (BEC): fraud that tricks employees into paying or changing payment details; voice-enabled variants add calls.

• Callback verification: calling back on a number you already have.

• Chargeback: a card dispute that reverses a charge.

• Consumer Sentinel: the FTC’s fraud-report database.

• Deepfake: synthetic media (audio, video, images) that falsely depicts someone.

• Deepfake detection: tools or methods for spotting synthetic media; none is perfect.

• Family emergency (“grandparent”) scam: a caller poses as a relative in trouble.

• Family safe word: a pre-agreed phrase to verify identity.

• Gift card scam: demanding payment in gift card numbers.

• IC3: the FBI’s Internet Crime Complaint Center.

• Liveness detection / liveness check: a test meant to show a real person is present.

• Malware: harmful software.

• Money mule: someone who moves stolen money, knowingly or not.

• One-time passcode: a single-use code sent to verify you; never share it.

• Out-of-band (independent-channel) verification: confirming through a different channel than the request.

• Peer-to-peer payment: app-based person-to-person payment.

• Phishing / smishing / vishing: fraud by email / text / phone.

• Robocall: an automated or prerecorded call.

• Social engineering: manipulating people into acting.

• Spoofing: falsifying caller or sender information.

• STIR/SHAKEN: an FCC-required framework that helps carriers verify caller-ID information, which reduces but does not eliminate illegal spoofing. theexchangesd

• Synthetic speech: machine-made speech.

• TCPA: the Telephone Consumer Protection Act.

• Text-to-speech vs. voice conversion: creating speech from text vs. changing one voice into another.

• Two-factor authentication (2FA): a second proof of identity beyond a password.

• Voice authentication (voiceprint): using your voice to log in.

• Voice cloning: software that imitates a specific person’s voice.

• Wire transfer: a fast bank-to-bank or service-to-service transfer, hard to reverse.

18. Frequently Asked Questions

What is an AI voice cloning scam?
It’s a fraud in which a criminal uses software to imitate a real person’s voice to pressure someone into paying or sharing information. The scripts, such as a family emergency or urgent business payment, are familiar. The cloned voice mainly makes the story more believable.

How does AI voice cloning work?
At a high level, software learns patterns from recordings of a person’s speech and then produces new speech that resembles them. Some systems start from text, others convert one voice into another. This article gives no steps, because the defense doesn’t depend on technical details.

Can scammers really clone my voice from a short clip?
The FTC has said a short audio clip may be enough. Published estimates of how much audio is needed vary, and quality differs by tool and recording, so treat any single number with caution. The practical lesson: assume any voice could be imitated and verify requests.

How do I know if a call is a voice clone?
You often can’t tell by listening. Look instead for warning signs: urgency, secrecy, a request for money or codes, and unusual payment methods. Hang up and call the person on a number you already have. That check works whether or not the voice is fake.

Can you tell a cloned voice by ear?
Not reliably. A 2023 peer-reviewed study found listeners spotted fakes about 73% of the time, even when they knew fakes were possible. Brief training helped only slightly. Real calls add stress and surprise, which make detection harder.

What is a family safe word?
It’s a word or phrase you agree on with family to confirm identity in an emergency. The FBI recommends creating one. It’s a useful layer, but people can forget it, so use it alongside a callback on a saved number.

Is voice cloning illegal?
Not automatically. The FBI says synthetic content isn’t inherently illegal but can facilitate crimes like fraud. Using a cloned voice to defraud, spoof caller ID with intent to harm, or place consent-less AI-voice robocalls can violate federal or state law. State laws vary.

What should I do if I get a call from a “relative” asking for money?
Hang up, then call the relative on a number you already have. If you can’t reach them, contact another relative or friend. Don’t send money, gift cards, or codes during the call. Tell someone you trust, and report the call at ReportFraud.ftc.gov.

What if I already sent money?
Contact your bank or payment provider immediately and tell them it was fraud. Then report to the FTC (ReportFraud.ftc.gov) and, if it happened online or electronically, to the FBI’s IC3 (ic3.gov). Speed matters because some payments can’t be reversed once they clear.

Can I get my money back after a voice cloning scam?
Sometimes, but it’s not guaranteed. It depends on the payment method, how fast you act, and the provider’s policy. Cards often have stronger dispute paths. Wires, gift cards, crypto, and cash are harder. Payments you were tricked into authorizing may not be covered by federal transfer rules.

Who do I report voice cloning scams to?
Report to the FTC at ReportFraud.ftc.gov, the FBI’s IC3 at ic3.gov for internet-enabled crimes, and your bank or payment provider. For spoofed or illegal calls, you can also file with the FCC. Your state attorney general and local police are additional options. Reports help investigators but may not recover money.

Can scammers fake caller ID?
Yes. The FCC describes spoofing as deliberately falsifying the information shown on your caller ID. A call that appears to come from a known name or local number proves nothing. Carriers use authentication tools like STIR/SHAKEN, but these reduce spoofing rather than eliminate it.

Are AI voice detectors reliable?
Not reliably. The FTC’s 2024 Voice Cloning Challenge found promising ideas but no single solution, and researchers have found automated detectors can fail when audio conditions change. Treat any detector result as one clue, never proof, and still verify through a trusted channel.

Can voice cloning bypass bank voice authentication?
It’s a recognized risk. The FBI has warned that AI audio could be used to gain access to bank accounts, and financial regulators encourage multiple authentication factors. Ask your bank what protects your account, use additional factors where offered, and don’t rely on voice alone.

How can I protect my elderly parents?
Agree on simple written rules: hang up, call back on a saved number, tell someone, and never pay by gift card, wire, or crypto on a phone request. Set up a family safe word, keep a contact list handy, and talk about it without shame.

How can businesses prevent voice-cloning fraud?
Use written procedures that don’t rely on recognizing a voice. Verify payment requests and bank-detail changes by calling a number on file, require dual approval for wires, allow no “urgent” exceptions, and train staff. Know your bank’s wire-recall contact before you need it.

How do I protect my voice online?
You can reduce exposure by tightening privacy settings and limiting public videos and voicemail greetings. Total removal isn’t realistic, and public audio is only one source. Focus most effort on verification habits, which protect you even when a voice sounds perfect.

What is vishing?
Vishing, or voice phishing, is fraud by phone that tries to get money, passwords, or codes. It long predates AI. Cloned voices are one way to make it more convincing. The same defense works: hang up and verify independently.

Are deepfake robocalls legal?
Generally not without consent. In February 2024 the FCC clarified that AI-generated voices count as “artificial” under the TCPA, so such robocalls need the same consent as other artificial-voice calls. Spoofing caller ID to defraud is separately unlawful. Exceptions and enforcement details are fact-specific.

What’s the difference between a deepfake and a voice clone?
A voice clone imitates someone’s voice. “Deepfake” is the broader term for synthetic media, including fake audio, video, and images. A voice clone is one kind of audio deepfake.

What are signs of a “virtual kidnapping” call?
The FBI describes extortion in which no one has actually been taken. Signs include calls not from the person’s own phone, callers who keep you on the line, refusal to let you check, and ransom demanded by wire. Try to reach the person directly or through someone near them, and contact authorities.

19. Sources

Items marked (secondary) summarize a primary document and should be checked against the original before publication. [VERIFY: replace secondary links with primary agency URLs where noted.]

Government and regulators

• FTC, Consumer Alert, “Scammers use AI to enhance their family emergency schemes,” March 2023. https://consumer.ftc.gov/consumer-alerts/2023/03/scammers-use-ai-enhance-their-family-emergency-schemes

• FTC, “What To Do if You Were Scammed.” https://consumer.ftc.gov/articles/what-do-if-you-were-scammed

• FTC, Consumer Alert, “A way to spot scams: how someone asks you to pay,” July 2026. https://consumer.ftc.gov/consumer-alerts/2026/07/way-spot-scams-how-someone-asks-you-pay

• FTC press release on 2025 imposter-scam data, June 2026. https://ftc.gov/news-events/news/press-releases/2026/06/ftc-data-show-people-reported-losing-3-point-5-billion-imposter-scams-2025

• FTC, testimony to the Joint Economic Committee on the rising scam economy. https://www.ftc.gov/system/files/ftc_gov/pdf/ftc-testimony-jec-hearing-on-the-rising-scam-economy.pdf

• FTC, “FTC Announces Winners of Voice Cloning Challenge,” April 8, 2024. https://search.ftc.gov/news-events/news/press-releases/2024/04/ftc-announces-winners-voice-cloning-challenge

• FTC, press release on the Impersonation Rule and proposed individual-impersonation extension, February 15, 2024. https://www.ftc.gov/node/82618

• Federal Register, final rule, Trade Regulation Rule on Impersonation of Government and Businesses (16 CFR Part 461), 2024. https://public-inspection.federalregister.gov/2024-04335.pdf

• FTC, announcement of Advance Notice of Proposed Rulemaking on impersonation and platforms, September 24, 2026. (secondary) https://www.mayerbrown.com/en/insights/publications/2026/09/ftc-seeks-public-comment-on-whether-to-update-rule-on-impersonation-of-government-and-businesses-to-address-platforms-role-in-promoting-impersonation-scams

• FBI IC3, PSA I-120324-PSA, “Criminals Use Generative Artificial Intelligence to Facilitate Financial Fraud,” December 3, 2024. https://www.ic3.gov/PSA/2024/PSA241203

• FBI IC3, PSA on impersonation of the IC3, July 20, 2026. https://www.ic3.gov/PSA/2026/PSA260720

• FBI IC3, 2025 Internet Crime Report, April 2026. https://www.ic3.gov/AnnualReport/Reports/2025_IC3Report.pdf

• FBI warning on senior-official impersonation (May 15, 2025, updated December 19, 2025). (secondary) https://www.biometricupdate.com/202512/ai-impersonation-attacks-against-us-officials-growing-more-sophisticated-fbi-warns

• FBI warning on altered “proof of life” photos, December 5, 2025. (secondary) https://www.yahoo.com/news/articles/fbi-warns-high-tech-virtual-164301082.html

• FBI Chicago, virtual-kidnapping warning signs. (secondary) https://nbcchicago.com/news/local/fbi-warns-of-virtual-kidnapping-scams-targeting-illinois-families/2775405

• FinCEN, Alert FIN-2024-Alert004, November 13, 2024. https://www.fincen.gov/sites/default/files/shared/FinCEN-Alert-DeepFakes-Alert508FINAL.pdf

• FCC, “FCC Makes AI-Generated Voices in Robocalls Illegal,” February 8, 2024. https://www.fcc.gov/document/fcc-makes-ai-generated-voices-robocalls-illegal

• FCC, “How Will Caller ID Authentication Help Consumers?” https://www.fcc.gov/node/228852

• FCC, Caller ID Spoofing. https://fcc.gov/spoofing

• FCC Consumer Complaint Center. https://consumercomplaints.fcc.gov

• FTC, IdentityTheft.gov. https://www.identitytheft.gov

• FDIC, consumer guide to credit and debit card billing issues (Regulation E and Regulation Z). https://archive.fdic.gov/view/fdic/6567/fdic_6567_DS1.pdf

• CFPB Regulation E FAQs (June 4, 2021). (secondary) https://dwt.com/blogs/financial-services-law-advisor/2021/06/cfpb-unauthorized-transfer-consumer-liability [VERIFY: link primary CFPB page.]

• National Elder Fraud Hotline. (secondary) https://www.loudounnow.com/opinion/columnists/virginia-seniors-reported-220-million-in-scam-losses-five-free-habits-can-help/article_2a4e139b-8a66-492d-bc2a-e999cc58161e.html [VERIFY on justice.gov.]

Statutes and proposals

• NO FAKES Act of 2026 (S. 4591 / H.R. 8915), reintroduction announced May 20, 2026. https://www.blackburn.senate.gov/2026/5/technology/blackburn-coons-salazar-dean-colleagues-introduce-revised-version-of-no-fakes-act

• NO FAKES Act committee action, June 18, 2026. (secondary) https://letsdatascience.com/news/no-fakes-act-advances-unanimously-out-of-senate-judiciary-co-553e714b

• Tennessee ELVIS Act, effective July 1, 2024. (secondary) https://www.reedsmith.com/articles/tennessee-efforts-protect-rop-against-genai-imitations-deepfakes/

Research and reporting

• Mai, Bray, Davies, Griffin, “Warning: Humans cannot reliably detect speech deepfakes,” PLOS ONE 18(8): e0285333 (2023). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10395974/

• Associated Press, coverage of remarks at a Federal Reserve conference, July 22, 2025. (secondary) https://www.local10.com/business/2025/07/22/openais-sam-altman-warns-of-ai-voice-fraud-crisis-in-banking

• Krebs on Security, coverage of FBI BEC advisory. (secondary) https://krebsonsecurity.com/?p=29648

• Team Gleason, voice banking for people with ALS. https://www.biospace.com/team-gleason-adds-and-quot-my-own-voice-and-quot-by-the-acapela-group-to-synthetic-voice-options-for-people-living-with-als

20. Conclusion

Voice cloning makes familiar-sounding scams more convincing, but the defense doesn’t depend on how real a voice sounds. Verifying through an independent channel works either way. Caller ID and detection tools are not proof. Payment method affects recovery, so act quickly and report, because speed and records both help. Laws, tactics, and data change, so confirm important details with the official sources above. Anyone can be targeted, and being targeted is not a personal failing.

This guide is a starting framework, not a final word. It is educational content and not legal, financial, or law-enforcement advice.