A North Korean scheme that replayed stolen webcam footage in fake meetings took more than $400 million. Lenders and title companies that move money by wire should take note.
Companies have spent years training staff to spot phishing emails and spoofed messages from executives. Video calls have mostly escaped from that suspicion. The FBI’s latest case suggests they shouldn’t.
On Nov. 12, the FBI and the Justice Department said they had disrupted “FaceStealer,” a North Korean operation. It lured targets into fake Zoom and Microsoft Teams meetings and played them secretly recorded webcam footage of people they knew. The FBI said the scheme stole more than $400 million in cryptocurrency over 18 months. Agents seized over 900 webcam recordings of victims.
The scheme is self-sustaining. Each person who joins a fake meeting is recorded, and that footage becomes the lure for the next target. The main targets so far have been crypto firms and open-source software developers. The method, though, would work against any company.
Hong Kong: $25 million on one call
Arup, the British engineering firm, has already learned what this costs. In January 2024, a finance employee in its Hong Kong office joined a video conference with people who looked and sounded like the company’s CFO and several colleagues. None of them were real. Fraudsters had built the participants from publicly available video and audio using deep-fake tools.
The employee followed their instructions and made 15 transfers totaling HK$200 million, about $25.6 million, to five bank accounts. The fraud came to light only when the employee later checked with headquarters.
What it means for mortgage and settlement firms
Mortgage lenders, title companies, settlement agents and banks are an obvious next target. Their staff regularly act on instructions from borrowers, attorneys, and executives they rarely meet in person. If that person appears convincingly on a video call, a request to change wiring instructions or speed up a disbursement can get past controls that were designed with email fraud in mind.
Firewalls and endpoint security don’t help when an employee is persuaded to approve a payment. Firms should consider:
- Out-of-band verification. Confirm payment instructions through a phone number or contact already on file, not one supplied during the call or in the request.
- Dual authorization. Require a second approver for transfers above a set threshold, and don’t let anyone on the original call fill that role.
- Account-ownership checks. Confirm that the receiving account belongs to the intended recipient before funds go out.
- Meeting hygiene. Join calls through the official app rather than an emailed link, especially when the contact is new or infrequent.
- Use a reliable third-party verification tool. These tools can instantly verify identity, bank information, and other details, so you are not fooled by scammers, often for a nominal transaction fee.
The broader point is structural. Confirming who someone is and approving a payment should be separate, independently controlled steps. A familiar face on a screen should satisfy neither.
The lesson for mortgage lenders
Just when you thought you had the whole wire fraud issue figured out, as usual the criminal elements become even more sophisticated. Train your staff, learn from the mistakes of others, and do not hesitate to outsource risk management when it comes to identity verification and electronic payment fraud.
In 2025, the FBI’s IC3 recorded nearly $21 billion in reported cybercrime losses. That is enough to make anyone sit up and take notice.