SDNY Indicts Three Men in $129 Million AI CEO Voice-Fraud Scheme

Federal prosecutors in Manhattan unsealed an indictment on Thursday charging three men with stealing more than $129 million by using artificial intelligence to impersonate chief executives and, in some cases, a lawyer at a major firm.

The indictment, filed in the U.S. District Court for the Southern District of New York, names Jacob Soall, Sharon Hazan and Guy Sheetrit. Prosecutors say the men ran the scheme between 2024 and 2026 and impersonated at least seven high-profile corporate executives worldwide. Each defendant faces seven counts, including wire fraud, aggravated identity theft and money laundering, exposing them to decades in prison if convicted.

How did the scheme work?

The trio targeted employees senior enough to authorize large wire transfers on behalf of their companies, according to the indictment. The defendants would persuade a target that help was needed on a confidential financial transaction, often a corporate acquisition.

They then invited the employee to a voice or video call. On that call, prosecutors say, AI did the heavy lifting.

"On those calls, the defendants and their co-conspirators would use AI to mimic the voice and at times the image of the impersonated CEO in order to convince the victim employee that the transaction was real."

After the CEO call, the victim was passed to an attorney — who was actually one of the men using AI to impersonate a real lawyer at a major law firm. "The impersonated attorney would ultimately direct the victim employee to wire a large sum of money in furtherance of that financial transaction to a particular bank account, which was in fact controlled by members of the conspiracy," prosecutors say. In sum, according to the government, "the purported corporate transactions were a sham."

The indictment does not identify the targeted companies or the impersonated law firms.

What amounts did each fraud bring in?

Prosecutors detailed three specific hits from 2025 alone:

• March 2025: more than $11.5 million from a United Kingdom-based fintech firm
• April 2025: more than $4.6 million from a Massachusetts-based mutual fund company
• June 2025: more than $5.2 million from a Florida-based pet food company

Each successful scam netted substantial sums, the government asserts, with the total across the conspiracy exceeding $129 million.

Who are the defendants?

Each man played a distinct role, prosecutors say:

• Jacob Soall, 41, a citizen of Spain and the United Kingdom, impersonated attorneys using a handwritten script filled with lawyerly vocabulary.
• Sharon Hazan, 54, of Israel, laundered the proceeds. Prosecutors quote him boasting that, in professional money laundering, he was "number one all over the world."
• Guy Sheetrit, 42, a citizen of Israel and the United Arab Emirates, also laundered the stolen cash.

Authorities arrested Soall in Spain in June, Hazan in Spain in August and Sheetrit in the United Kingdom in September. Their extradition requests remain pending.

What did the U.S. Attorney say?

U.S. Attorney Jamie McDonald for the Southern District of New York framed the case as a warning about AI-enabled crime, using President Donald Trump's preferred term, "super intelligence":

"As alleged, Jacob Soall, Sharon Hazan, and Guy Sheetrit abused groundbreaking super intelligence technology to impersonate real people and deceive corporate executives into sending them and their co-conspirators approximately $129 million."

McDonald added: "I encourage individuals and the business community to consider how to best protect themselves from the emerging threat of SI-enabled crime."

Why does this matter for corporate counsel?

The indictment is the latest AI-focused prosecution out of Manhattan's federal court. Earlier this year, in the same court, a North Carolina man pleaded guilty to creating hundreds of thousands of AI-generated songs and using bots to stream them, generating more than $8 million in fraudulent royalties. He received an 18-month prison sentence for one count of conspiracy to commit wire fraud.

For in-house lawyers and transaction teams, the SDNY case is a concrete demonstration that voice and video are no longer reliable authentication factors in deal execution, and that callback verification and out-of-band confirmation protocols must now cover not only counterparty executives but also the outside counsel who supposedly bless the wire instructions.

The bottom line: AI-generated authority is now an indictment-level fraud vector, and payment controls that assume a real voice on the line are out of date.