Deepfake

What it is

Deepfakes can put someone's face on another body, sync lips to scripted audio, or recreate a voice. Some are entertainment; others are weapons for fraud — fake CEO video calls, fabricated scandals, or emergency pleas that never happened.

You do not need Hollywood tools anymore. Short public clips are often enough for a usable fake, which is why skepticism and verification matter more than ever.

How it works

1. Source material — Photos, videos, or voice samples are collected from social posts, interviews, or leaks.
2. Generation — Software maps the face or voice onto new content.
3. Delivery — The deepfake arrives as a call, video meeting, social post, or attachment.
4. Ask — The script pushes a payment, secret, login, or reputation hit.
5. Amplify — Sometimes the fake is shared widely to pressure the victim.
6. Deepfakes power many AI impersonation and romance/pig-butchering overlays.

Red flags

• Video or audio of someone asking for money or secrets under extreme urgency
• Odd lighting, blurred ears/hairlines, unnatural blinking, or mismatched room audio
• Refusal to switch to a known channel or answer a private shared-memory question
• A live call that feels pre-recorded or laggy in strange ways
• Content that appears on a platform the person rarely uses
• Pressure to act before you can verify with others

What to do if you’re targeted

• **Verify offline** before you send money, share codes, or reshare damaging content.
• **Call the person** on a number you already have — do not trust the inbound deepfake channel alone.
• **If you have been impersonated:** document the content, report it to the platform, and warn contacts.
• **If you paid:** contact your bank and report to ReportFraud.ftc.gov / IC3.gov (U.S.).
• **Limit public samples** where practical, and use MFA on accounts attackers might target next.
• **Optional check-in:** a free identity scan if the scam also harvested personal data.

Related terms

• AI impersonation
• AI voice clone
• Social engineering