A phone call from your CEO. A follow-up from a lawyer at a well-known law firm. A request to wire millions of dollars, and fast. U.S. authorities say that exact playbook, powered by artificial intelligence, was used in an alleged AI CEO impersonation scheme that caused more than $129 million in losses.
Key facts
- U.S. authorities announced on November 5, 2026 that Jacob Soall, 41, Sharon Hazan, 54, and Guy Sheetrit, 42, were charged in an alleged corporate impersonation scheme linked to more than $129 million in losses.
- According to reporting on the indictment, the defendants allegedly posed as company CEOs and as attorneys from prominent law firms, using AI-generated or manipulated voices and video to make wire transfer instructions look genuine.
- Reported victims include a Missouri hospital system that lost approximately $4.83 million and a UK fintech company that lost approximately $11.58 million.
- The three face federal fraud, identity theft and money laundering charges, have not been convicted, and were reportedly arrested in Europe while awaiting extradition.
What Authorities Say About the AI CEO Impersonation Scheme
The case, covered by ABC News and the Associated Press, centers on a simple idea with a high-tech twist. Fraudsters allegedly convinced employees that senior leaders and outside counsel were directing urgent payments.
The charges are allegations. None of the three defendants has been convicted, and they are presumed innocent unless proven guilty in court.
Who is charged
Authorities named three defendants: Jacob Soall, 41, Sharon Hazan, 54, and Guy Sheetrit, 42. Each faces federal charges that, according to reporting, include fraud-related counts, identity theft and money laundering.
Reporting indicates the trio were arrested in Europe. They were reportedly awaiting extradition to the United States at the time of the announcement.
How the alleged scheme worked
According to reporting on the indictment, the operation relied on layered impersonation. The goal was to make a payment request feel like it came from someone an employee already trusted and was used to obeying.
- Email impersonation of CEOs. The defendants allegedly sent messages posing as chief executives of targeted companies.
- AI-generated or manipulated voice and video. Investigators say synthetic or altered audio and video were used to make instructions seem authentic.
- Fake attorneys. The defendants allegedly impersonated lawyers from prominent law firms, adding a layer of legal authority to the request.
- Large wire transfers. Employees were allegedly persuaded to send substantial sums to accounts controlled by the conspirators.
Reports state that no impersonated CEO or law firm is accused of taking part in the scheme. Those named were victims of identity misuse, not participants.
Known Victims of the Alleged AI Impersonation Fraud
Court reporting identified two specific victims so far. Their losses show how a single successful call can translate into a multimillion-dollar hit.
- A Missouri hospital system: approximately $4.83 million lost.
- A UK fintech company: approximately $11.58 million lost.
Those two figures account for only a portion of the total. Additional coverage from Courthouse News and Newser describes the broader pattern of targeted companies, though the full victim list has not been detailed in the sources reviewed.
Where the Money Allegedly Went
Investigators say the stolen funds moved through accounts controlled by the conspirators. Some of the proceeds were allegedly converted into cryptocurrency.
That detail matters. Once wire transfers are layered through multiple accounts and swapped into digital assets, recovery becomes far harder for victims and banks. It is also why money laundering charges often accompany impersonation cases like this one.
What the $129 Million Figure Really Means
The $129 million number is the reported aggregate loss attributed to the alleged operation. It is not an amount proven to have been personally received by any one defendant.
That distinction is important for accurate reporting. Total victim losses, money moved through intermediary accounts, and funds each person allegedly kept are separate questions. Prosecutors will have to establish those details if the case proceeds to trial.
Why AI Impersonation of Executives Is So Dangerous
Business email compromise has been a costly problem for years. Adding synthetic voice and video raises the stakes because the traditional safeguard, "I recognized the voice," no longer holds up.
Employees are trained to respond to authority, especially when a message looks urgent and confidential. A familiar voice from a boss, backed by a note from an attorney, can override instincts that would normally flag a strange request.
Three features make this style of fraud effective:
- Urgency. Requests are framed as time-sensitive deals or legal matters.
- Secrecy. Targets are told not to discuss the transaction, which blocks natural checks.
- Borrowed credibility. Names of real executives and real law firms lend false legitimacy.
How Companies Can Defend Against an AI CEO Impersonation Scheme
No single tool stops deepfake-assisted fraud. Strong process is the best defense, because process does not depend on whether an employee can spot a fake.
Build verification into every large payment
- Require a second approver for wire transfers above a set threshold.
- Confirm payment changes through a separate, known channel, such as a phone number already on file, not one supplied in the request.
- Treat any instruction to bypass normal approval as a red flag, no matter who appears to give it.
- Use a pre-agreed code phrase or challenge question for executive requests involving money.
Train staff on synthetic media
- Show finance, legal and executive assistant teams examples of cloned voices and manipulated video.
- Make clear that nobody will be punished for pausing a payment to verify it.
- Run simulated impersonation drills, just as many firms do for phishing.
Move fast when something looks wrong
Speed matters after a fraudulent wire. Contact your bank immediately to request a recall, and report the incident to law enforcement. Victims in the United States can file a report with the FBI Internet Crime Complaint Center, and individuals can also use the FTC fraud reporting portal.
Part of a Wider Pattern of AI-Driven Fraud
This case is the latest example of criminals using synthetic media to exploit trust. Earlier reporting on the Raj Chakraborty AI voice scam showed how cloned voices can pressure targets into acting without checking. Another report on an Elon Musk deepfake scam arrest highlighted how a trusted public face can be used to lure victims.
The cross-border angle here also echoes other prosecutions. Our coverage of a extradited crypto scam suspect shows how long the path from arrest abroad to a U.S. courtroom can be.
What Happens Next
The next major step is extradition. Until the defendants appear in a U.S. court, the case will move slowly, and many details of the indictment may stay limited to what prosecutors have disclosed.
Key questions remain. How many companies were targeted? How much was recovered? And how were the AI-generated voices and video produced and deployed? Answers should emerge as court filings become public.
Conclusion: Verify Before You Wire
The alleged AI CEO impersonation scheme shows how convincing a fake boss or fake lawyer can be when technology polishes the lie. A Missouri hospital system and a UK fintech firm are among those reportedly hit, and the total reported loss tops $129 million.
The defendants are charged, not convicted, and the facts will be tested in court. For every business, the takeaway is practical. Set up call-back verification, require dual approval for large transfers, and train your team to pause when a request feels rushed. If you have been targeted, report it to the FBI IC3 and your bank right away, and share this article with your finance team today.
Frequently Asked Questions
Who was charged in the AI CEO impersonation scheme?
U.S. authorities charged Jacob Soall, 41, Sharon Hazan, 54, and Guy Sheetrit, 42. They face federal fraud, identity theft and money laundering charges. They have not been convicted and are presumed innocent unless proven guilty.
How much money was lost in the alleged AI impersonation fraud?
Reporting attributes more than $129 million in aggregate losses to the alleged operation. That figure is not an amount proven to have been received by any single defendant. Named victims include a Missouri hospital system (about $4.83 million) and a UK fintech company (about $11.58 million).
How did the alleged scheme use AI?
According to reporting on the indictment, the defendants used AI-generated or manipulated voices and video to make instructions appear to come from company CEOs. They also allegedly impersonated attorneys from prominent law firms. The goal was to persuade employees to send large wire transfers.
Were the real CEOs or law firms involved?
No. Reporting states that no impersonated CEO or law firm is accused of taking part in the scheme. Their identities were allegedly used without involvement to make the requests look legitimate.
How can a company protect itself from deepfake CEO fraud?
Require a second approver for large wires and verify payment requests through a known phone number, not contact details in the request. Train staff to recognize synthetic voice and video, and allow them to pause payments without penalty. If fraud occurs, contact your bank immediately and report it to the FBI IC3.
