theaisafemode.com
AI Safe Mode

Deepfake Bosses Allegedly Tricked Firms Out of $129M

An empty meeting room seen through a glass wall; illustrative
Illustrative photograph only, not an event image. Photo by Crew via Wikimedia Commons; CC0 1.0. See LICENSES.md.

US prosecutors say three people used AI deepfakes of company bosses and lawyers to push staff into wiring about $129M. How to stop a fake boss request.

US prosecutors in New York have charged three people over an alleged scheme that used AI deepfakes of company chief executives and lawyers to persuade staff to send about $129 million by wire transfer.[1] [2] The defendants are presumed innocent, and lawyers for at least one of them have denied involvement.[1]

What happened

According to the reports, the alleged scheme ran from July 2024 to February 2026. Finance and accounting employees were contacted by people posing as their own senior executives or as outside lawyers handling a confidential deal. AI-generated voices and video were reportedly used to make the requests feel real.[1] [2]

The targets reportedly included hospital systems and a bank. Prosecutors say the money was moved through accounts in several countries and some of it was converted into cryptocurrency. Arrests were reported in Spain and the UK, and the charges include wire fraud, aggravated identity theft and money laundering.[1] [2]

Why it matters

The trick is old: a “boss” asks for an urgent, secret payment. What has changed is how convincing the boss can look and sound. A familiar face on a video call used to be a strong signal. It is now only a starting point.

The pattern works because it stacks pressure. The request is urgent, it is confidential, and it appears to come from someone senior enough that questioning it feels risky. A lawyer joins to make it sound official. None of that proves the request is genuine.

The short version

If a payment request arrives with secrecy and urgency attached, slow down. Verify it through a channel the requester did not choose.

How to stay safe

  • Call back on a known number. Use the number in your company directory or contacts, not one in the email, chat or call.
  • Use a two-person rule. Large or unusual transfers should need a second person to approve them.
  • Treat secrecy as a warning sign. A real executive can wait while you check through normal channels.
  • Agree a check question. Teams can set a private phrase or question for money requests that a deepfake would not know.
  • Report fast. If money has gone, call your bank straight away; a quick report gives the best chance of stopping a transfer.

Follow AI Safe Mode

Protect yourself with Privaku

See the post on YouTube Watch and follow AI Safe Mode on YouTube @AISafeMode and Instagram @aisafemode.

References

  1. The Guardian, Nov 6, 2026. Report on the charges and the alleged method. Read the report ↗
  2. CyberScoop, Nov 6, 2026. Report on the indictment and the deepfake impersonations. Read the report ↗
Facts come only from the published reports above and are attributed to them. AI Safe Mode doesn't name companies, products or people involved in incidents. Spotted an error? Email [email protected] and it will be reviewed and corrected. See the corrections policy.