DOJ Charges Three Men in $129 Million AI CEO-Impersonation Scheme

Admin


Three foreign nationals have been charged in New York with using artificial intelligence to impersonate corporate executives and attorneys in a fraud scheme that allegedly stole at least $129 million from hospitals, banks and other businesses around the world. According to an indictment unsealed on Thursday, November 5, in the Southern District of New York, the defendants are Jacob Soall, 41, of Marbella, Spain and the United Kingdom; Sharon Hazan, 54, of Israel; and Guy Sheetrit, 42, of Israel and the United Arab Emirates.

The scheme ran from at least July 2024 through at least February 2026 and targeted thousands of companies. Prosecutors say the conspirators typically contacted employees in accounting roles who had authority to move large sums, posing as the company's CEO and claiming a confidential deal was underway. On follow-up phone or video calls, they used AI to mimic the executive's voice and sometimes even their appearance, then brought in supposed attorneys from major law firms to vouch for the transfer. Among the victims were a Missouri hospital system that lost $4.83 million, a California hospital system that lost $2.97 million, a Rhode Island bank that lost about $2.7 million, a Massachusetts mutual fund that lost more than $4.6 million, a Florida pet food company that lost about $5 million, and a UK technology company that lost more than $11 million. The money moved through accounts in countries including the US, Spain, Portugal, Greece, the United Arab Emirates, Singapore, Austria, Slovakia and China.

Each defendant faces seven counts: conspiracy to commit wire fraud, wire fraud, two counts of aggravated identity theft, money laundering conspiracy and two counts of money laundering. Wire fraud and money laundering each carry maximum sentences of 20 years, while each aggravated identity theft count carries two years. Soall was arrested in Spain in June, Hazan in Spain in August and Sheetrit in the United Kingdom in September; extradition requests are pending, and all three are presumed innocent unless proven guilty. Notably, the Justice Department's press release never says "AI" — it calls the technology "SI," short for "super intelligence," throughout.

Why it matters

This is the first major case where deepfakes are the method itself, not a garnish. The fraud worked because the voice on the phone sounded exactly like the boss, the face on the video call looked like the boss, and a "lawyer" from a real firm backed the story up. No firewall stops that — only a process does, and the companies that lost money evidently did not have one: a second channel to verify a wire instruction before it moves.

Expect that to change quickly. Callback verification for any transfer ordered by voice or video is about to become standard finance hygiene, and insurers and auditors will start asking whether it exists. The case also sets a marker for how prosecutors will talk about AI in court — "SI" in the charging documents suggests they already see the technology as something distinct from a tool, and the sentences they are seeking show they intend to treat AI-enabled fraud as seriously as the fraud itself.

Source: Gizmodo — DOJ Says Deepfaked CEOs and Lawyers Helped Steal $129 Million From Hospitals, Banks, and Others (2026-11-07). Summary rewritten by VCTRIA.

Post a Comment