A woman in Ohio gets a call on a Tuesday morning. Her grandson's voice β and it really does sound like her grandson's voice β tells her he's in a holding cell in Tijuana, he can't talk long, please don't tell his mother, the lawyer needs $9,500 in gift cards by the end of the day to keep him from being charged. She drives to three different drugstores so no single clerk gets suspicious. She mails the card numbers to an address in Texas. Her grandson is, of course, at work in Cincinnati. The first anyone in the family hears about the call is when her daughter notices the bank statement six weeks later.
Stories like this one are now the central financial-safety problem in American families with an aging parent. The Federal Trade Commission's most recent Protecting Older Consumers report, released in December 2025 covering 2024 data, puts reported fraud losses by adults 60 and older at $2.4 billion β a fourfold increase since 2020. Because most fraud goes unreported, the FTC's own estimate of the true cost runs between $10.1 billion and $81.5 billion. Two-thirds of the reported dollars come from individual losses of $100,000 or more, the kind that wipe out a retirement account in an afternoon.
That is the number that matters to a family. The median older victim doesn't lose $50. They lose a meaningful share of what they spent forty years building.
This guide is for the adult child, spouse, or sibling who is trying to make a parent or partner harder to scam without taking away their autonomy. It covers the six categories of scam doing the most damage in 2026, why older adults are targeted, the single hardest sub-case (a loved one being scammed who won't believe you), what the AI voice-cloning inflection point means for grandparent calls, and a short list of things any family can do in one weekend that meaningfully reduces the risk.
Why scammers target older adults
The polite reason is that older adults have savings. The honest reasons are more specific, and naming them helps with the work:
- Accumulated wealth. A 78-year-old who has paid off a house and saved through a working career is, statistically, a much larger target than her 38-year-old daughter. Scammers know this.
- Time and isolation. A retired widow who lives alone can take a 40-minute phone call without anyone walking into the room. A scammer's script depends on getting twenty uninterrupted minutes; that's harder to engineer with a working-age adult who has a co-worker at the next desk.
- Generational politeness with authority. Many older Americans were raised to be cooperative with anyone who introduced themselves as Officer, Doctor, Agent, or Manager. Hanging up on a "federal agent" feels rude even when the rational mind knows it shouldn't.
- Unfamiliarity with newer payment rails. Gift cards, cryptocurrency, and same-day wire transfers entered everyday life recently enough that the gut-level "this is weird" reflex hasn't fully formed in someone who has been writing paper checks for sixty years.
- In some cases, declining executive function. This one matters but is the easiest to misuse. Most older adults are cognitively intact and many are sharper than their kids; treating "she's old" as a synonym for "she's confused" insults the parent and misses the actual mechanism. Where mild cognitive impairment is present, it tends to affect specifically the ability to recognize manipulation in real time β exactly the function a scam call is engineered to overwhelm.
A useful frame: the scammer is not winning because the parent is gullible. The scammer is winning because the call is professionally produced, the script has been A/B-tested on thousands of previous targets, and the parent is one person against a small business.
The six scams doing the most damage
1. Government impersonation: Medicare, Social Security, IRS, "officer with a warrant."
How it works: an automated or live call says there's a problem with the target's Medicare account, Social Security number, or tax filing. The voice gets serious. There may be a fake "case number" and a transfer to a fake "agent." The threat is some combination of: your benefits will be suspended, your SSN has been linked to a crime in Texas, there's a warrant out for your arrest, agents are on the way unless you can clear it up right now. Resolution requires either reading your Social Security number aloud to confirm identity, or paying a fine in gift cards or wire transfer.
The tell: real federal agencies do not call to threaten arrest. Medicare does not phone members asking to verify a number. The IRS sends letters. The Social Security Administration sends letters. No legitimate government agency takes payment in gift cards, ever.
What to do: hang up. If the call mentions a specific case number that sounds plausible, look up the agency's published phone number yourself β not a number the caller gave you β and call back.
2. Tech-support scams.
How it works: a pop-up appears on the parent's computer, often with a loud alarm sound, claiming the device has a virus or has been "locked by Microsoft." A phone number is displayed. When the parent calls, a "technician" asks for remote access to fix the problem, then either installs real malware, charges hundreds of dollars for an unnecessary service, or escalates into a bank-account "refund" scam where the victim is convinced to wire money back.
The tell: Microsoft, Apple, and Google never display a phone number in a security warning. They do not call customers. A real warning closes when you close the browser tab.
What to do: shut the browser. If that doesn't work, hold the power button until the device turns off. Then turn it back on. Older adults reported $159 million in losses to tech-support scams in 2024 alone β almost all of it preventable by walking the parent through the "just turn it off" muscle memory once.
3. The grandparent / family-emergency scam.
How it works: a call comes in late at night or first thing in the morning. A voice says "Grandma?" When the grandparent answers with a name β "Michael?" β the scammer becomes Michael. The story is always an emergency in a place that's hard to verify: a car accident in another state, an arrest in Mexico, a hospitalization after a bar fight. The plot has two consistent features: it requires money fast, and it requires secrecy, usually framed as please don't tell Mom.
The tell: the urgency and the secrecy together. Real emergencies generate phone calls to parents, not requests to hide things from them. Real lawyers do not take gift cards. Real bail bondsmen are bonded by the state and accept normal payment methods.
What to do: hang up and call the family member directly on the number you already have for them. If you can't reach them, call their parent or spouse. The scam survives only the first thirty seconds; any independent verification ends it.
We will come back to this one separately because AI voice cloning has changed the math.
4. Romance scams.
How it works: a months-long online relationship, often initiated on a social-media platform or dating site. The scammer is patient, attentive, emotionally fluent. Eventually a financial emergency materializes β a medical bill while traveling, a customs hold on a shipment of gold, a need to fly home to the United States to finally meet in person. Money is requested in increments. The relationship continues. More money is requested.
The tell: the person has never been on video in real time. The story for why not is always plausible (overseas military deployment, oil rig contract, deep-sea fishing crew, surgeon working in a war zone). The financial requests start small and grow.
What to do: this is the hardest scam to confront a parent about, because the loss isn't only money β it's also the relationship, which the parent has experienced as real. Lead with the financial pattern, not with "he's not real." A useful question: would you be willing to wait 48 hours before sending anything, and call me first? Waiting is what the scammer cannot tolerate. The relationship dies the moment the cash flow stops.
5. Sweepstakes, lottery, and "you've won" scams.
How it works: a notification arrives β by phone, mail, or email β saying the recipient has won a large prize, often from Publishers Clearing House, a foreign lottery, or a sweepstakes they don't remember entering. To collect, they need to pay "taxes," "processing fees," or "customs duties" up front, usually by gift card or wire.
The tell: real prize organizations deduct taxes from the winnings. They do not ask the winner to send money before receiving any. You also cannot win a lottery you did not enter.
What to do: throw the letter out, hang up the call, delete the email. If the parent has been responding for a while and small payments have already gone out, the address and phone number are now on a "suckers list" that gets sold among scammers. Expect more attempts. This is a case for changing the phone number.
6. Gift cards and wire transfers as the payment method.
This is less a separate scam than a universal red flag that cuts across all of the above. No legitimate organization on earth takes payment in iTunes cards, Google Play cards, Target cards, or Steam cards. No real government agency, utility, bank, attorney, hospital, or bail bondsman accepts cryptocurrency or same-day wire transfer as a way to resolve an emergency. If the demand for payment specifies gift cards or wire transfer, the call is a scam β full stop, no exceptions, no nuance.
This is the single most useful sentence to plant in an aging parent's head: anyone who asks for payment in gift cards is a criminal. It does not matter how convincing they sound or what title they claimed. The payment method itself is the diagnosis.
The AI voice-cloning inflection point
For about thirty years, the grandparent scam relied on the grandparent being half-asleep, the connection being bad, and the script being good enough to bridge the gap between "Grandma?" and a wire transfer. The defense was simple: ask a question only the real grandchild would know. What's the name of your dog? Where did we go last Christmas? The scammer, who knew nothing about the family, would stumble.
That defense stopped working in 2024.
Commercial AI voice-synthesis tools released in the last two years can clone a recognizable voice from as little as three to thirty seconds of source audio. The source can be a TikTok video, a voicemail greeting, an Instagram reel, a YouTube comment, a wedding toast posted on Facebook. The output is good enough β over a cellphone connection β to fool the speaker's own mother. The FBI issued a public-service announcement in 2024 and updated it in 2025; the AARP Fraud Watch Network has been tracking a sharp increase in reports through 2025 and into 2026.
This changes the family-emergency scam from a phone-acting performance into a content-production pipeline. A scammer can scrape a teenager's TikTok, generate a thirty-second cloned voice clip, and use it inside an interactive call where the teenager's "voice" begs grandma for bail money. The grandmother's question β Sweetheart, is that really you? β gets answered with what sounds exactly like her grandchild saying yes, Grandma, it's me.
There is one defense that still works, and it is mechanical rather than auditory: a family code word.
A family code word is a short phrase β two or three random words, ideally β that the real family members know and a cloned voice does not. It is not a pet's name, a street address, a school mascot, or anything else that could be found on social media. It is something arbitrary: blue stapler birthday, Thursday pineapple lighthouse, the title of the book the family read together on a vacation in 1997. It is established once, in person or over a known-secure channel, and never written down anywhere a scraper can find it.
When a call comes in claiming a family emergency, the answer is not to ask what's wrong, where they are, or how much money is needed. The answer is to say, with deliberate calm: Before we go any further, what's our code word? The real family member will know. The cloned voice will not.
This is the single most actionable defensive move a family can make in 2026. Pick a phrase tonight. Tell the parent, the spouse, the grandchildren, the siblings. Repeat it at the next family dinner. Then trust it.
The hardest sub-case: when the loved one is being scammed and won't believe you
This is the situation that breaks families. A parent has been in an online relationship for nine months. A widow has been sending checks to someone who calls himself a U.S. Marshal. A father has been buying Apple gift cards twice a week and reading the numbers to a "Microsoft technician" he calls every morning at 10 a.m. The adult children have noticed. They've named it. They've shown the parent articles. The parent has thanked them for caring and continued sending money.
A few things to know.
Shame doesn't work. The instinct is to confront, often in front of other family members, often with anger that comes from real fear about the money. This almost always entrenches the parent's position. Scam victims have usually been told by the scammer that their family won't understand, that the children are jealous, that everyone is trying to keep them isolated. Confrontation confirms the scammer's framing. The parent retreats further into the relationship that is, in their experience, the only one that respects them.
Seizing the phone doesn't work either. Beyond the autonomy violation, it's tactically poor: the parent has the phone number written down somewhere, or they buy a new prepaid phone, or they use a neighbor's. The scam will continue. What disappears is the family's visibility into it.
What does work, sometimes, is patient listening. Ask the parent to walk you through the story. Don't argue. Ask questions about the parts that don't add up β not gotcha questions, but real ones. How did he get your number? When you sent the gift cards, why did he need them in five-hundred-dollar increments? You said the money is for him to fly here β when does the flight land? The goal is not to win the argument. It's to let the parent assemble the contradictions themselves. Sometimes they do. Sometimes they don't, and the next steps become unavoidable.
Report it externally. The AARP Fraud Watch Network Helpline at 877-908-3360 is staffed by trained specialists who handle exactly this conversation, including how to talk to a parent who is in denial. They will also help walk through reporting to the FTC at reportfraud.ftc.gov or by phone at 877-FTC-HELP (877-382-4357). If the parent is over 60 and the losses are recurring or large, Adult Protective Services in their county is the right next call β find the local number via the Eldercare Locator at 800-677-1116 or eldercare.acl.gov.
Consider a financial power of attorney. This is the conversation most families avoid until after the second large loss. A durable financial power of attorney lets a trusted family member step in when the parent is being defrauded, without waiting for a court to declare incapacity. Pair it with a "trusted contact" designation at the parent's bank and brokerage β a FINRA-regulated process that lets the institution call you when they spot an unusual withdrawal. Most banks will not act on a trusted-contact alert by freezing the account, but the phone call alone often gives the family the early warning that's been missing.
None of these steps is a clean fix. The honest version is that a parent who is being romanced or threatened by a scammer over months is in a relationship the family is competing with, and competing on the family's actual terms β patience, presence, real conversation β is the only way the competition gets won.
What to do this week
A short checklist a family can complete in a single weekend:
- Pick a family code word. Two or three random words, not derivable from social media. Tell every member of the family in person or on a call you already trust. Don't write it down where a scraper or phone thief can find it. Use it the next time anyone calls claiming an emergency.
- Register the parent's phone number with the National Do Not Call Registry at donotcall.gov or by calling 1-888-382-1222. It will not stop scammers (who don't follow the law anyway) but it cuts the volume of legitimate telemarketing enough that the scam calls stop blending in.
- Install a scam-call blocker on the parent's phone. Nomorobo, RoboKiller, Truecaller, and the built-in spam filters from the major carriers all help. None is perfect; together they catch most of the worst offenders.
- Freeze the parent's credit at all three bureaus β Equifax, Experian, and TransUnion. It's free, it takes about fifteen minutes per bureau online, and it prevents new accounts from being opened in the parent's name. Unfreezing is easy when they actually need new credit.
- Set up real-time bank alerts for any transaction over a threshold the family agrees on β $500 is a common starting point. Most banks let alerts go to the parent and to one designated family member.
- Designate a trusted contact at the parent's bank and brokerage. Under FINRA's trusted-contact-person rule, this gives the institution a permitted channel to call a family member when they spot a suspicious withdrawal. It does not give the trusted contact authority over the account β it gives them an early-warning phone call.
- Have the durable financial power of attorney conversation, even if you don't act on it. A POA signed when the parent is fully capable and stored where the family can find it is one of the most valuable single pieces of paper an adult child can have. It does not transfer any power until it is exercised; it just exists, ready, if it ever needs to.
- Bookmark four phone numbers in the parent's phone. AARP Fraud Watch (877-908-3360), FTC (877-FTC-HELP), Eldercare Locator (800-677-1116), and your own number labeled "call me before sending money." The last one is the one that matters most.
A script for the conversation
The opening conversation with a parent about scam risk goes better when it isn't framed as I'm worried you're going to fall for something. A version that lands more cleanly:
"Mom, this isn't about you β these scams are getting so good they're catching everyone. There's a thing now where they can fake my voice from a TikTok. I want us to have a code word so that if you ever get a call that sounds like me asking for money, you know how to check. Can we pick one right now?"
Three things this script does. It names the threat as external, not as the parent's vulnerability. It puts the asking adult in the same boat β they can fake my voice too β rather than positioning the parent as the weak link. And it ends with a specific small action, not a vague be careful out there.
Most parents say yes. The few who push back ("I'd never fall for that") still usually agree to the code word, because the code word costs nothing to set up and proves nothing about them. Once it exists, the harder conversations get easier, because the family has already had one practical conversation about scam safety that wasn't a fight.
The bottom line
Older Americans lost somewhere between $2.4 billion and $81.5 billion to fraud in 2024, depending on how you count. The number isn't going down in 2025 or 2026, because AI-generated voice, AI-generated video, and AI-generated text have collapsed the cost of running a convincing scam from something only a skilled human could do to something a teenager with a laptop can do at scale.
The thing that hasn't changed is what protects a family. A parent who has a code word, a frozen credit report, a trusted contact at the bank, a phone with scam-call filtering on, and an adult child whose number is the first one they call before sending money β that parent is a much harder target than the one across the street with none of those.
None of this stops the calls. The calls will keep coming. What it does is reduce the chance that any single call connects with a panicked, isolated, hurried decision, which is the only condition under which the scam works.
The scammer's whole business depends on the family not being in the room. The work, simply, is making the family more in the room.