Anyone can end up on the wrong end of a video call that is not what it seems. Deepfake scams use artificial intelligence to copy a person's face, voice, and movements to fool careful people. Here is how they work and how to protect your money.
What is a deepfake, really?
A deepfake is a video, photo, or audio recording created by AI to look and sound like a real person. The software learns how a face moves, how a voice rises and falls, and how a person pauses, then puts those patterns into someone else's mouth. The same technology powers entertainment and creative work. In a scammer's hands, it is a tool for one goal: making a lie look like the truth.
The scams scammers build on deepfakes
Scammers use deepfakes in a few patterns that keep coming up:
- Fake CEO requests. An employee gets a video call or voice message from a "senior executive" asking for a large transfer, gift cards, or login details. The pressure is intense because the boss is on camera.
- Family emergency calls. A parent hears a cloned voice that sounds exactly like a child, begging for money after a supposed accident or arrest. This builds on classic family emergency scams with a far more convincing voice.
- Fake celebrity endorsements. Clips of well-known people appear to promote investment apps or crypto schemes. Investment scam red flags look different when a trusted face is attached.
- Romance scams with a face. A dating app match offers video calls that look real but are generated from stolen images. The romance scams playbook now includes believable video.
Why your eyes are not a reliable lie detector
Here is the honest limit: you cannot reliably spot a good deepfake with your eyes. Early fakes had blurred edges and odd shadows. Modern ones do not. Detection tools exist, but they are not perfect, and neither are people.
Voice is easier to fake than most people expect. A short sample from a social video is enough to train some cloning tools. Video calls also hide flaws through compression, lighting, and low resolution.
So stop treating "can I spot it?" as the defense. The winning move is not looking harder. Check the facts separately.
A real request can always survive a real check.
The real defense: verify through an independent channel
When a call, video, or message asks for money or sensitive information, slow down and confirm it separately:
- Hang up and call back a known number. Not the one the caller gave you — the one you already have saved, or the official number from a website you trust.
- Use a separate channel. If they contacted you by video, respond by phone. A scammer running one fake channel rarely controls your real ones.
- Agree on a codeword. Choose a private phrase with family and trusted colleagues. Use it whenever money is involved, no matter how real the caller sounds.
- Ask a question only the real person could answer. This works best when the answer is not posted anywhere online. Combined with a codeword, it is a strong check.
Be careful what you share with AI tools
Some apps offer to clone your voice, animate your photos, or turn a selfie into a video. That can be fun, but there is a cost — the software needs many samples of your voice and face to learn them, and you may not know where they go afterward. Before you upload a voice recording or face scan, check what the tool does with your data and whether you can delete everything. The less raw material you hand out, the less a scammer has to work with.
How to handle a suspicious call
If something feels off, you do not have to be polite. End the call, write down what happened, and verify before acting. Do not pay by gift card, wire transfer, or cryptocurrency, since those payments are almost impossible to reverse. Then report it — in the United States, file a report with the FBI Internet Crime Complaint Center at IC3 and tell the FTC.
If someone uses your face or voice without permission
Discovering a fake clip of yourself is disturbing, but it can be handled in steps: save links and screenshots as evidence before anything is taken down, report the content to the platform (most sites ban deceptive media), report the misuse to the FTC and to IC3, and tell the people closest to you so they know a fake version of you exists and can ignore it.
Quick answers
How can I tell if a video is a deepfake? With a well-made one, you often cannot. Look for odd movement around the face, unnatural blinking, or audio that does not match the lips. But never rely on your eyes alone — verify through a separate channel and use a codeword.
Can scammers clone my voice from a short video? Yes. Some tools can work from a very short sample of clear speech. Avoid posting long, unedited recordings of your voice and avoid samples on unknown apps.
What should I do if a family member calls asking for money? Treat it as real until confirmed. End the call, call back on a number you already have, and use your codeword.
Where do I report a deepfake scam? Report it to the FTC, and file a complaint with the FBI Internet Crime Complaint Center at IC3 if money was lost or fraud is suspected. Also report the fake video to the platform.
Sources and further reading
Written by Hassan Arshad, founder of UsefulOrbit. Last updated August 31, 2026.