How AI is Supercharging Fraud
The scams are nothing new, the online “sweetheart” who needs a loan, the "fraud department" calling about your account, the grandchild in trouble. What used to protect us was that the fakes were flawed. A scam email was riddled with typos. A caller impersonating your son didn't really sound that much like your son. A fake cryptocurrency investment site looked haphazardly cobbled together. Those tells are now gone. Artificial intelligence writes flawless, personalized messages in seconds, clones a voice from a few seconds of audio, and generates convincing fake websites, documents, and even live video on demand. The con is the same, the level of polish is what’s new.
Last week I wrote about the quiet ways we leave our financial lives exposed, the accounts we link, the statements we leave in the mailbox. This is the other side of that coin: what happens when a fraudster targets you directly on a personal level now that the fraud has gotten dramatically more convincing. The numbers are already sobering. In its most recent annual report, the FBI's Internet Crime Complaint Center broke out "AI" as its own fraud category for the first time, logging more than 22,000 complaints and nearly a billion dollars in losses tied specifically to artificial intelligence. And because most fraud goes unreported, even those grim figures are almost certainly a fraction of the truth.
The scams to know
A few patterns account for most of the damage.
The voice-clone emergency. This is the most emotionally devastating of them all. You get a frantic call: your child, your grandchild, your spouse, in a car accident or under arrest, needing money wired immediately. The voice is real, or sounds exactly like it, because it was cloned from a social media clip, a voicemail greeting, or a video posted online. All it takes is a few seconds of recorded audio to generate the deepfake voice and create a moment of panic.
The "your account is compromised" text or call. Bank-impersonation scams have exploded, and AI makes them seamless. You get a text that looks exactly like a fraud alert from your bank, followed by a call from a calm, professional "representative" offering to help you move your money somewhere safe, an account they control. Real banks never do this.
The long con, now automated. In "pig butchering" schemes, a scammer builds a relationship over weeks or months, often starting with a friendly wrong-number text or a dating app match, before introducing a “can't miss” investment opportunity or asking for a loan, often in crypto. AI has industrialized this. One criminal can now run hundreds of these patient, convincing conversations at once, each with a fabricated persona and, increasingly, deepfaked photos and video to prove the person is "real."
The deepfake video call. Once the stuff of science fiction, this is now a documented corporate nightmare. In the most striking case, an employee at a global engineering firm wired roughly $25 million after joining a video call with what appeared to be the company's chief financial officer and several colleagues. Every person on that call was a deepfake. The same technology that fooled a trained finance professional is getting cheaper and easier by the month.
The thread connecting all of them isn't just technical wizardry. It's psychology: manufactured urgency, a demand for secrecy, and a push toward a payment you can't claw back, a wire, gift cards, or cryptocurrency. The technology just makes the lie look and sound flawless.
Who gets hit hardest
It's worth saying plainly: older Americans bear the brunt of this. People 60 and over reported losing roughly twice as much to AI-driven scams as those just a decade younger. This isn’t necessarily because older folks are more gullible. Part of it is that scammers deliberately target retirees with larger pools of savings. Part of it is that this technology moved so fast that the warnings haven't caught up. If you have parents or grandparents, the single most important thing you can do after reading this is to forward it to them, or better yet, call them and have a conversation.
How to actually protect yourself
Here's the reassuring part, you do not need to outmatch the criminals with technology or learn to spot a perfect fake, which experts now say is nearly impossible anyway. The defenses are refreshingly human and low-tech, and they work precisely because they don't depend on your ability to detect the forgery.
Verify the request, not the voice. Stop trying to judge whether a voice or a face is real; that battle is lost. Instead, when any urgent request for money or information comes in, hang up and make contact through a channel you already trust. Call the person back on their known number. Text them. Call the bank using the number on the back of your card, never the one the caller gives you. A real emergency survives a five-minute pause. A scam rarely does, because urgency is the scammer's most important weapon, and refusing to be rushed disarms it.
Agree on a family code word. This is the single most effective defense against the voice-clone emergency, and it costs nothing. Pick a word or phrase that only your immediate family knows, something random and unguessable, and share it with them in person (don’t put this in a text message or email). When a panicked voice calls asking for money, ask for the word. A cloned voice won't have it.
Guard your voice and face. The raw material for these fakes comes from what we post. You don't have to vanish from the internet, but think twice about public videos where you speak at length, lock down your social media privacy settings, and be wary of "fun" apps that ask you to record your voice.
Treat irreversible payments as a giant red flag. Almost no legitimate transaction requires payment by wire transfer, gift card, or cryptocurrency on the spot. When someone insists on one of these, especially in a hurry, that alone is almost a sure sign of a scam.
The bottom line
Artificial intelligence has taken the oldest tricks in the book and made them look and sound perfect. That's the genuinely unsettling part: the instincts that protected us for generations, trusting a familiar voice or even believing our own eyes, no longer offer the protection they once did. But the flip side is that the defenses haven't really changed, they've just become more important. Slow down. Verify through a channel you trust. Agree on a code word. Be deeply suspicious of urgency and of any payment you can't reverse. None of it requires you to understand the technology. It just requires you to pause at the exact moment the scam is engineered to make you panic. In an age when you can no longer believe everything you see and hear, the most valuable thing you can do is give yourself the sixty seconds it takes to verify.
Disclosure
Convivia Financial LLC is a registered investment advisor. This article is for general informational purposes only and reflects the author's opinions as of the date of publication, which are subject to change without notice. Nothing herein constitutes investment, legal, or tax advice. All investing involves risk, including loss of principal. Information from third-party sources is believed reliable but has not been independently verified. Registration as an investment advisor does not imply any particular level of skill or training. For more information about the Firm, including fees and conflicts of interest, please refer to our Form ADV Part 2A available upon request.
