Hong Kong Firm Loses $25.6 Million in Elaborate Deepfake Video Conference Scam
An employee of a multinational firm in Hong Kong was defrauded of $25.6 million (HK$200 million) in an elaborate deepfake scam, the South China Morning Post reported. Scammers posed as the firm’s chief financial officer (CFO) in a group video conference call and ordered money transfers from an employee at the finance department of the firm, according to the police. This is being highlighted as first of its first-of-its-kind case in the city.
Wow Deepfake AI capability is some wild and concerning stuff
— Chris Combs (iterative design enjoyer) (@DrChrisCombs) February 5, 2024
A Hong Kong employee was fooled by a Deepfake impersonation of his company's CFO on a virtual call (with multiple other Deepfaked employees as well) to transfer $25M USD
We're used to being suspicious of emails but… pic.twitter.com/mpJF5ZP4hK
How the elaborate deepfake scam unfolded
The scam started when the employee received a suspicious email purportedly from the company’s UK-based CFO. The employee initially brushed off the mail as it seemed like a phishing message and the email instructed the employee to execute a secret transaction. However, the employee was convinced about the message when he got on a Zoom video conference call with the CFO and 15 other employees of the company. The people in attendance looked and sounded exactly like the colleagues he recognized.
That wasn't actually me on a Zoom call, it was a malicious AI clone built from TV interviews, claims PR A Binance PR exec says crooks created a deep-fake "AI hologram" of him to scam cryptocurrency projects via Zoom video calls.… https://t.co/MgCM4cDhSi
— Java Wales (@JavaWales) August 23, 2022
Believing everyone else on the call was real, the worker agreed to remit a total of HK$200 million via 15 transfers to five different Hong Kong bank accounts, the police said in the report.
The police officials also said that the scammers asked the victim to introduce himself but did not interact with him during the meeting. Further, the fake images on the screen mainly gave orders and the call ended abruptly. The scammers stayed in touch with the victim through instant messaging platforms, emails, and follow-up one-on-one video calls.
However, everyone present on the video calls, except the victim, was fake. The scammers used advanced deepfake technology to turn publicly available video and other footage of the people into convincing deepfake versions of the meeting’s participants including the company’s CFO. The entire scam lasted for about a week from the time the employee was contacted till he realized it was a scam. It was uncovered when the victim made an official inquiry with the company’s headquarters.
The acting senior superintendent Baron Chan Shun-ching said scammers approached other employees as well using the same multi-person video call tactic. About two to three employees in total had been approached by the scammers, but they did not provide full information on their encounters, the police said in the report.
The first-of-its-kind scam in Hong Kong
Superintendent Chan said this was the first case that involved multiple people as in previous cases, the scam victims were tricked only by one-on-one video calls. “This time, in a multi-person video conference, it turns out that everyone you see is fake,” said the Hong Kong police official, as quoted by the SCMP.
Chan further advised the public that to avoid being scammed by deepfake video calls, people should ask the callers questions and request them to move their heads to determine the authenticity. Further, people should naturally grow suspicious the moment money is requested.
Fake deepfake calls of President Joe Biden circulate ahead of primary
— WION (@WIONews) January 23, 2024
How to be safe from such hoax calls? @Mohammed11Saleh tells you more
Watch more on https://t.co/AXC5qRuO3J pic.twitter.com/2KXfnuaDX8
However, this is one of the several deepfake AI scams that have unfolded in the past couple of years. Most recently, pop star Taylor Swift was a victim of a deepfake scam in which scammers created and leaked fake explicit images of her on the internet. The incident nudged the EU negotiators to criminalize some online violence against women, including graphic AI-generated material, per Politico.
Sexually explicit deepfake images of Taylor Swift have led lawmakers to propose the DEFIANCE Act of 2024. These images have sparked public conversation on the uses & dangers of AI in the world of sexual exploitation. Today, on Safer Internet Day, what are your thoughts? pic.twitter.com/DZGD6zqjud
— Operation Underground Railroad (@OURrescue) February 7, 2024
In the modern world, not just data but people's voices, faces, bodies and more are all present on social media and the internet. All of these can be used to create deepfake AI images as the technology has become super sophisticated. Thus, lawmakers and tech companies across the globe are trying to look for ways to prevent frauds.