📞 One Word Stops This Scam Cold

Good morning. My mother almost fell for a couple of phone scams years before AI existed. Today's lead looks at the modern, AI-powered version of that same call, the one that can now sound exactly like your grandson. Then a two-minute way to check any suspicious message before you click anything.

Ask ChatGPT or Gemini to record a bedtime story in your own voice for a grandchild, and it does it convincingly: warm, familiar, unmistakably you. The exact same technology, needing as little as three seconds of your grandson's voice pulled from a social post, now powers the modern grandparent scam: a panicked call from someone who sounds exactly like him, saying he has been arrested and needs bail money right now.

McAfee's global survey of 7,000 people found that one in four had run into an AI voice scam themselves or knew someone who had. Among the people actually targeted, most who lost money lost somewhere between a few hundred dollars and well into five figures.

I take this personally. Years ago, long before any of this existed, my mother nearly fell for a couple of phone scams. We caught it before she sent a dime. That was a stranger reading off a script. Today's version can sound like one of us.

The catch: there is no tell left to listen for. The voice is not close enough to fool Grandma. It is identical.

The fix costs nothing and takes one conversation. Pick a family code word nobody outside the house knows. If a panicked caller cannot say it, hang up and call that person back on the number you already have.

Go further: The FTC's own guidance on voice cloning ◆ How family emergency scams actually work

FROM OUR PARTNER

AI is moving fast and most people are falling behind. 

The Rundown AI keeps you ahead of the curve. 

It's a free AI newsletter that keeps you up-to-date on the latest AI news, and teaches you how to apply it in just 5 minutes a day.

Plus, complete the quiz after signing up and they’ll recommend the best AI tools, guides, and courses — tailored to your needs.

Sign up to start learning.

AI TUTORIAL

The old advice was to look for bad grammar or an odd turn of phrase. That tell is gone. AI writes a perfectly fluent scam message now, spelling and all, so the check has to happen somewhere else.

1. Paste the full message into ChatGPT, Claude, or Gemini, unedited, exactly as you received it.

2. Ask it to list every red flag: urgency, a request for money or a gift card, a link that doesn't match the sender's real domain, an unusual ask.

3. Do not click the link yourself. Ask AI to describe where it actually points, based on the visible URL text, without opening it.

4. Ask who the sender claims to be, then check that name or number against a contact you already have saved, not one in the message.

5. If anything mentions a family member in trouble, stop. Call that person directly on a number you already know is theirs.

If it does not work: AI can sometimes wave off a real scam or flag a genuine message as suspicious, especially if the message is short on detail. Treat its answer as a second opinion, not a verdict, and still call the person directly if anything asks for money.

What this doesn't fix: AI cannot verify who actually sent something or confirm a number belongs to who it claims. That confirmation still has to happen through a channel you already trust, not one supplied by the message itself.

Go further: The FTC's full guide to spotting phishing

ASK A FAMOUS MIND
💼 What would Steve Jobs say about quitting a comfortable job at 55 and starting over?
Ask him yourself in Ask a Famous Mind. Always free. 👉 Try it here

NEWS YOU CAN USE
This Matters at Home
Short, practical summaries to help you stay updated on recent AI news—and what it means for everyday life.

🚨 Fake AI Videos Are Telling Seniors They Can Skip Property Taxes. A viral AI-generated video falsely claims Congress passed a law letting seniors freeze property taxes, then asks for a Social Security number to "file." No such program exists.

👨‍👧 A Dad Who Studies AI Says Don't Ban It, Teach Both Ways. Economist Daniel Susskind says kids should learn skills with AI and without it, so they can tell when the tech is wrong. Reading and math still matter.

🩺 Patients Are Bringing ChatGPT Into the Doctor's Office. More patients arrive prepped with ChatGPT's take on their symptoms and labs. Doctors like the head start, as long as confident AI guesses don't replace real medical judgment.

REAL WORLD
How We Used AI This Week
Real people, real uses—here’s how AI is quietly making life better, one small win at a time.

🌸 Jenna P., London, England. Decades into gardening, she spotted a flowering plant at Dunrobin Castle she could not name, the kind of moment that used to mean posting a photo and waiting for other gardeners to weigh in online. This time she uploaded the picture to Gemini and ChatGPT, and both identified it immediately as Cleome hassleriana, spider flower. ChatGPT explained which features it used to make the call. She confirmed it later and now wants to grow one herself.

👉 What's the last thing you asked AI at home? Tell us here. One question, 30 seconds.

READER POLL

If today's lead has you thinking about how scammers get your family's information in the first place, Incogni tackles the supply side. It scans hundreds of data broker sites that quietly buy and sell your name, address, and phone number, then requests removal on an ongoing basis, not just once. Less of your information floating around online means fewer places a scammer can start. Start here.

PASS IT ON
Know someone who would like this newsletter? Send them here.

Missed an issue? Read the archive.