Scammers are using deepfakes, 'urgent' family emergencies to get your money. Losses could reach $16B by late 2026. Spot the signs before it's too late

And online fraudsters are cashing in. More than 4.2 million fraud reports have been filed since 2020, resulting in more than $50.5 billion in losses, “with a growing portion stemming from deepfake scams,” according to the FBI (1). But these numbers are likely much higher since many victims of online scams don’t report them to law enforcement.

Financial losses from deepfake and other AI-generated scams alone are projected to reach $40 billion in the U.S. alone by 2027, according to Deloitte’s Center for Financial Services (2).

And the problem is only going to get worse. Cybersecurity firm DeepStrike estimates an increase from 500,000 deepfakes in 2023 to 8 million in 2025. “This is consistent with a growth rate where the volume of deepfake videos increases by 900% annually. This isn’t linear growth; it’s a viral proliferation that outpaces nearly every other cyber threat,” according to DeepStrike (3). Here’s how they are rising in use so quickly and how to protect yourself.

The rise of an almost-perfect scam

Deepfakes aren’t just being used to trick corporations and banks. The same techniques are being used to up the ante on the classic ‘grandchild scam,’ in which fraudsters use deepfakes to mimic family members in distress — faking an emergency that requires an urgent transfer of funds to cover a hospital bill or bail, or maybe even ransom for a kidnapping.

This scam is particularly effective — and particularly nefarious — because it bypasses logic and targets emotion, triggering a sense of urgency. And that’s the point, since it causes even the most rational person to act before verifying.

“By the end of 2026, so we’re talking just over a year, it’s going to be the majority of the way scams are done,” Roger Grimes, a cybersecurity expert with Clearwater-based security awareness company KnowBe4, told Tampa Bay 28, during an investigation into how AI deepfake scams are targeting families (4).

“Hackers are doing these fake kidnappings. They’re calling grandparents to say that your grandchild has been in an accident. They need $5,000 to get out of trouble,” Grimes said (4).

AI tools are already cheap, accessible and easy to use, meaning nearly anyone with a social media presence could be vulnerable.

Audio deepfakes are one of the most widespread forms of deepfake attacks. “With as little as three seconds of genuine audio, AI voice generators can mimic someone’s voice with up to 95% accuracy,” according to cybersecurity firm McAfee (5).

For video deepfakes, scammers can take advantage of publicly available images and videos scraped from sources like Instagram, TikTok and YouTube. And human detection rates are just 24.5% for high-quality video, according to DeepStrike (3).

The aftermath of a deepfake scam can be devastating on many levels. You likely won’t be able to get your money back, since banks may refuse to reimburse losses for transactions that were ‘authorized.’ More than two-thirds (77%) of people targeted by an AI voice scam lost money as a result, with 36% losing between $500 and $3,000, according to a McAfee survey (6).

Plus, if you shared sensitive data with the scammers, they could potentially use that for identity theft — and commit more fraud, such as taking out loans in your name (and hurting your credit score). Victims may also feel ashamed and confused, triggering anxiety, depression and distrust.

How to protect yourself from deepfakes

While state and federal lawmakers are attempting to enact laws to protect consumers against deepfake scams, it’s hard for lawmakers to keep up with rapidly advancing technology that currently doesn’t have a lot of adequate guardrails.

This, unfortunately, puts much of the onus on consumers to protect themselves. In this environment, you’ll need to assume that hearing a loved one’s voice or seeing them on a video is no longer proof of identity.

But there are a few red flags to look out for, as well as verification habits that could protect your finances.

“Even if it’s a deepfake AI, they’ve got to still scam you,” Grimes told Tampa Bay 28. “If it’s an unexpected message asking you to do something you’ve never done before, research it, even if it sounds and looks like somebody you know” (4).

In videos, look for signs such as jitter (brief flickers), unnatural eye movement and inconsistent lighting and shadows. With phone calls, the voice may sound exactly like your family member, but an out-of-character request and sense of urgency should raise your Spidey senses. Listen for red flags such as unnatural interruptions, unusual pacing or repetitive phrasing.