Staff Reports
The Southern District of New York unsealed a dramatic indictment this week accusing three foreign nationals of using AI-enabled impersonation to steal more than $129 million from companies around the world. The accused — Jacob Soall, Sharon Hazan, and Guy Sheetrit — were arrested abroad and extradition requests are pending. The Justice Department even used the new term “super intelligence,” and announced a Super Intelligence Task Force the same day. This isn’t science fiction. It’s a criminal playbook gone high-tech.
What prosecutors say happened
According to the indictment, the scheme was simple and cruelly effective. Criminals cloned executives’ voices and images, then called employees and ordered urgent, secret wire transfers. They followed up by pretending to be outside lawyers who confirmed the instructions. Prosecutors found a handwritten “Lawyer Call” script used to silence questions. Stolen money was moved into accounts, converted to cryptocurrency, and laundered across borders. United States Attorney Jamie McDonald warned that “SI technology offers tremendous potential. But the public should recognize that there are some who seek to use this emerging technology to pursue criminal goals.” FBI Assistant Director in Charge James C. Barnacle, Jr. noted this shows how advanced tools can be weaponized for large-scale fraud.
How good people made a bad decision
There’s nothing mystical here. The attackers relied on human pressure. A worker hears their boss saying “handle this quietly,” a lawyer confirms it, and someone who wants to look competent hits “send.” That failure is not a technology problem alone. It’s a workplace culture problem. Companies that let titles and urgency override basic checks set the table for criminals. The FBI and fraud monitors have been warning: voice or video alone is not proof of identity. Yet too many firms still take the shortcut because the boss wanted it done yesterday.
“Super intelligence” and the policy gap
Yes, the Justice Department chose the phrase “super intelligence,” and it rolled out a task force to match. Fine — name it whatever sounds serious. Names don’t stop bad actors. Enforcement and clear rules do. We need tougher penalties for cross-border AI-enabled fraud, faster international cooperation on extradition, and mandatory safeguards for high-value transfers. Banks and crypto platforms that accept laundered funds should face real consequences. Meanwhile, companies should require multi-person sign-off, independent contact verification, and limits on how quickly large transfers can be made. Empower employees to pause and verify without fearing a pop quiz from the boss later.
Final takeaway
This SDNY indictment should be a wake-up call, not a headline people nod at and forget. AI impersonation, deepfake voice scams, and business-email compromise are now blended into one lethal tool for thieves. The practical fix starts inside companies: train people, tighten procedures, and let employees say plainly, “I’m checking first.” That five-word sentence may be the cheapest and most effective fraud prevention a business can buy — and the most patriotic act an employee can offer in a world where nothing is quite what it seems.
